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
Richter -log A0 table magnitude: log-domain component average · case 01
Unequal component amplitudes bias ML upward.
Richter -log A0 table magnitude: log-domain component average · case 02
Unequal component amplitudes bias ML upward.
Richter -log A0 table magnitude: log-domain component average · case 03
Unequal component amplitudes bias ML upward.
Richter -log A0 table magnitude: log-domain component average · case 04
Unequal component amplitudes bias ML upward.
Richter -log A0 table magnitude: log-domain component average · case 05
Unequal component amplitudes bias ML upward.
Richter -log A0 table magnitude: dead channel exclusion · case 01
A dead horizontal channel drags ML down by 1.5 units.
Richter -log A0 table magnitude: dead channel exclusion · case 02
A dead horizontal channel drags ML down by 1.5 units.
Richter -log A0 table magnitude: dead channel exclusion · case 03
A dead horizontal channel drags ML down by 1.5 units.
Richter -log A0 table magnitude: dead channel exclusion · case 04
A dead horizontal channel drags ML down by 1.5 units.
Richter -log A0 table magnitude: dead channel exclusion · case 05
A dead horizontal channel drags ML down by 1.5 units.
Richter -log A0 table magnitude: interpolation anchor node · case 01
Between table nodes the distance correction moves in the wrong direction; only node distances are right.
Richter -log A0 table magnitude: interpolation anchor node · case 02
Between table nodes the distance correction moves in the wrong direction; only node distances are right.
Richter -log A0 table magnitude: interpolation anchor node · case 03
Between table nodes the distance correction moves in the wrong direction; only node distances are right.
Richter -log A0 table magnitude: interpolation anchor node · case 04
Between table nodes the distance correction moves in the wrong direction; only node distances are right.
Richter -log A0 table magnitude: interpolation anchor node · case 05
Between table nodes the distance correction moves in the wrong direction; only node distances are right.
Regional conversion to Mw: ML low branch boundary · case 01
ML 3.5 converts to Mw 3.4.
Regional conversion to Mw: ML low branch boundary · case 02
ML 3.5 converts to Mw 3.4.
Regional conversion to Mw: ML low branch boundary · case 03
ML 3.5 converts to Mw 3.4.
Regional conversion to Mw: ML low branch boundary · case 04
ML 3.5 converts to Mw 3.4.
Regional conversion to Mw: ML low branch boundary · case 05
ML 3.5 converts to Mw 3.4.
Regional conversion to Mw: Ms branch boundary · case 01
Ms 6.1 converts with the large-magnitude relation.
Regional conversion to Mw: Ms branch boundary · case 02
Ms 6.1 converts with the large-magnitude relation.
Regional conversion to Mw: Ms branch boundary · case 03
Ms 6.1 converts with the large-magnitude relation.
Regional conversion to Mw: Ms branch boundary · case 04
Ms 6.1 converts with the large-magnitude relation.
Regional conversion to Mw: Ms branch boundary · case 05
Ms 6.1 converts with the large-magnitude relation.
Regional conversion to Mw: ML saturation · case 01
Saturated ML 6.5 is passed through as Mw 6.5.
Regional conversion to Mw: ML saturation · case 02
Saturated ML 6.5 is passed through as Mw 6.5.
Regional conversion to Mw: ML saturation · case 03
Saturated ML 6.5 is passed through as Mw 6.5.
Regional conversion to Mw: ML saturation · case 04
Saturated ML 6.5 is passed through as Mw 6.5.
Regional conversion to Mw: ML saturation · case 05
Saturated ML 6.5 is passed through as Mw 6.5.
Regional conversion to Mw: unknown type handling · case 01
Md and lower-case ml values are published as Mw unchanged.
Regional conversion to Mw: unknown type handling · case 02
Md and lower-case ml values are published as Mw unchanged.
Regional conversion to Mw: unknown type handling · case 03
Md and lower-case ml values are published as Mw unchanged.
Regional conversion to Mw: unknown type handling · case 04
Md and lower-case ml values are published as Mw unchanged.
Regional conversion to Mw: unknown type handling · case 05
Md and lower-case ml values are published as Mw unchanged.
Scalar moment from moment tensor: off-diagonal multiplicity · case 01
Strike-slip and oblique mechanisms get Mw too small.
Scalar moment from moment tensor: off-diagonal multiplicity · case 02
Strike-slip and oblique mechanisms get Mw too small.
Scalar moment from moment tensor: off-diagonal multiplicity · case 03
Strike-slip and oblique mechanisms get Mw too small.
Scalar moment from moment tensor: off-diagonal multiplicity · case 04
Strike-slip and oblique mechanisms get Mw too small.
Scalar moment from moment tensor: off-diagonal multiplicity · case 05
Strike-slip and oblique mechanisms get Mw too small.
Scalar moment from moment tensor: Frobenius half factor · case 01
Mw is 0.1 units too high for every tensor.
Scalar moment from moment tensor: Frobenius half factor · case 02
Mw is 0.1 units too high for every tensor.
Scalar moment from moment tensor: Frobenius half factor · case 03
Mw is 0.1 units too high for every tensor.
Scalar moment from moment tensor: Frobenius half factor · case 04
Mw is 0.1 units too high for every tensor.
Scalar moment from moment tensor: Frobenius half factor · case 05
Mw is 0.1 units too high for every tensor.
Scalar moment from moment tensor: dyne-cm to newton-metre · case 01
Every CMT Mw is 4.67 units high.
Scalar moment from moment tensor: dyne-cm to newton-metre · case 02
Every CMT Mw is 4.67 units high.
Scalar moment from moment tensor: dyne-cm to newton-metre · case 03
Every CMT Mw is 4.67 units high.
Scalar moment from moment tensor: dyne-cm to newton-metre · case 04
Every CMT Mw is 4.67 units high.
Scalar moment from moment tensor: dyne-cm to newton-metre · case 05
Every CMT Mw is 4.67 units high.
Scalar moment from moment tensor: isotropic trace test · case 01
Implosive (negative trace) sources are labelled deviatoric.
Scalar moment from moment tensor: isotropic trace test · case 02
Implosive (negative trace) sources are labelled deviatoric.
Scalar moment from moment tensor: isotropic trace test · case 03
Implosive (negative trace) sources are labelled deviatoric.
Scalar moment from moment tensor: isotropic trace test · case 04
Implosive (negative trace) sources are labelled deviatoric.
Scalar moment from moment tensor: isotropic trace test · case 05
Implosive (negative trace) sources are labelled deviatoric.
Station magnitude correction update: correction sign · case 01
Applying the corrections doubles the station bias instead of removing it.
Station magnitude correction update: correction sign · case 02
Applying the corrections doubles the station bias instead of removing it.
Station magnitude correction update: correction sign · case 03
Applying the corrections doubles the station bias instead of removing it.
Station magnitude correction update: correction sign · case 04
Applying the corrections doubles the station bias instead of removing it.
Station magnitude correction update: correction sign · case 05
Applying the corrections doubles the station bias instead of removing it.
Station magnitude correction update: residual outlier bound · case 01
Residuals of exactly one magnitude unit are discarded.
Station magnitude correction update: residual outlier bound · case 02
Residuals of exactly one magnitude unit are discarded.
Station magnitude correction update: residual outlier bound · case 03
Residuals of exactly one magnitude unit are discarded.
Station magnitude correction update: residual outlier bound · case 04
Residuals of exactly one magnitude unit are discarded.
Station magnitude correction update: residual outlier bound · case 05
Residuals of exactly one magnitude unit are discarded.
Station magnitude correction update: minimum residual count · case 01
Stations with exactly the minimum number of residuals get no correction.
Station magnitude correction update: minimum residual count · case 02
Stations with exactly the minimum number of residuals get no correction.
Station magnitude correction update: minimum residual count · case 03
Stations with exactly the minimum number of residuals get no correction.
Station magnitude correction update: minimum residual count · case 04
Stations with exactly the minimum number of residuals get no correction.
Station magnitude correction update: minimum residual count · case 05
Stations with exactly the minimum number of residuals get no correction.
Abe tsunami magnitude: far-field gauge limit · case 01
A tide gauge at exactly 100 km is ignored.
Abe tsunami magnitude: far-field gauge limit · case 02
A tide gauge at exactly 100 km is ignored.
Abe tsunami magnitude: far-field gauge limit · case 03
A tide gauge at exactly 100 km is ignored.
Abe tsunami magnitude: far-field gauge limit · case 04
A tide gauge at exactly 100 km is ignored.
Abe tsunami magnitude: far-field gauge limit · case 05
A tide gauge at exactly 100 km is ignored.
Abe tsunami magnitude: single amplitude · case 01
Mt is 0.3 units high at every gauge.
Abe tsunami magnitude: single amplitude · case 02
Mt is 0.3 units high at every gauge.
Abe tsunami magnitude: single amplitude · case 03
Mt is 0.3 units high at every gauge.
Abe tsunami magnitude: single amplitude · case 04
Mt is 0.3 units high at every gauge.
Abe tsunami magnitude: single amplitude · case 05
Mt is 0.3 units high at every gauge.
Abe tsunami magnitude: distance term · case 01
Mt is about two units low.
Abe tsunami magnitude: distance term · case 02
Mt is about two units low.
Abe tsunami magnitude: distance term · case 03
Mt is about two units low.
Abe tsunami magnitude: distance term · case 04
Mt is about two units low.
Abe tsunami magnitude: distance term · case 05
Mt is about two units low.
Abe tsunami magnitude: gauge averaging · case 01
Events recorded by near-field gauges have their Mt diluted.
Abe tsunami magnitude: gauge averaging · case 02
Events recorded by near-field gauges have their Mt diluted.
Abe tsunami magnitude: gauge averaging · case 03
Events recorded by near-field gauges have their Mt diluted.
Abe tsunami magnitude: gauge averaging · case 04
Events recorded by near-field gauges have their Mt diluted.
Abe tsunami magnitude: gauge averaging · case 05
Events recorded by near-field gauges have their Mt diluted.
S-P nomogram local magnitude: arrival order · case 01
Swapped phase picks still produce a magnitude.
S-P nomogram local magnitude: arrival order · case 02
Swapped phase picks still produce a magnitude.
S-P nomogram local magnitude: arrival order · case 03
Swapped phase picks still produce a magnitude.
S-P nomogram local magnitude: arrival order · case 04
Swapped phase picks still produce a magnitude.
S-P nomogram local magnitude: arrival order · case 05
Swapped phase picks still produce a magnitude.
S-P nomogram local magnitude: distance proxy factor · case 01
Magnitudes are 2.7 units low.
S-P nomogram local magnitude: distance proxy factor · case 02
Magnitudes are 2.7 units low.
S-P nomogram local magnitude: distance proxy factor · case 03
Magnitudes are 2.7 units low.
S-P nomogram local magnitude: distance proxy factor · case 04
Magnitudes are 2.7 units low.
S-P nomogram local magnitude: distance proxy factor · case 05
Magnitudes are 2.7 units low.
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 ↗