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
Moment magnitude from fault dimensions: rigidity units · case 01
Moment is a thousand times too small and Mw is 2 units low.
Moment magnitude from fault dimensions: rigidity units · case 02
Moment is a thousand times too small and Mw is 2 units low.
Moment magnitude from fault dimensions: rigidity units · case 03
Moment is a thousand times too small and Mw is 2 units low.
Moment magnitude from fault dimensions: rigidity units · case 04
Moment is a thousand times too small and Mw is 2 units low.
Moment magnitude from fault dimensions: rigidity units · case 05
Moment is a thousand times too small and Mw is 2 units low.
Moment magnitude from fault dimensions: SI moment constant · case 01
Every moment magnitude is 4.7 units low.
Moment magnitude from fault dimensions: SI moment constant · case 02
Every moment magnitude is 4.7 units low.
Moment magnitude from fault dimensions: SI moment constant · case 03
Every moment magnitude is 4.7 units low.
Moment magnitude from fault dimensions: SI moment constant · case 04
Every moment magnitude is 4.7 units low.
Moment magnitude from fault dimensions: SI moment constant · case 05
Every moment magnitude is 4.7 units low.
Moment magnitude from fault dimensions: magnitude scale factor · case 01
Large ruptures come out with implausible magnitudes above 15.
Moment magnitude from fault dimensions: magnitude scale factor · case 02
Large ruptures come out with implausible magnitudes above 15.
Moment magnitude from fault dimensions: magnitude scale factor · case 03
Large ruptures come out with implausible magnitudes above 15.
Moment magnitude from fault dimensions: magnitude scale factor · case 04
Large ruptures come out with implausible magnitudes above 15.
Moment magnitude from fault dimensions: magnitude scale factor · case 05
Large ruptures come out with implausible magnitudes above 15.
Equivalent magnitude of a swarm: energy summation · case 01
A swarm with one large event is reported far smaller than that event alone.
Equivalent magnitude of a swarm: energy summation · case 02
A swarm with one large event is reported far smaller than that event alone.
Equivalent magnitude of a swarm: energy summation · case 03
A swarm with one large event is reported far smaller than that event alone.
Equivalent magnitude of a swarm: energy summation · case 04
A swarm with one large event is reported far smaller than that event alone.
Equivalent magnitude of a swarm: energy summation · case 05
A swarm with one large event is reported far smaller than that event alone.
Equivalent magnitude of a swarm: energy inversion · case 01
The equivalent magnitude of a single M5 event is reported as M11.25.
Equivalent magnitude of a swarm: energy inversion · case 02
The equivalent magnitude of a single M5 event is reported as M11.25.
Equivalent magnitude of a swarm: energy inversion · case 03
The equivalent magnitude of a single M5 event is reported as M11.25.
Equivalent magnitude of a swarm: energy inversion · case 04
The equivalent magnitude of a single M5 event is reported as M11.25.
Equivalent magnitude of a swarm: energy inversion · case 05
The equivalent magnitude of a single M5 event is reported as M11.25.
Equivalent magnitude of a swarm: exponential base · case 01
Totals are many orders of magnitude too small.
Equivalent magnitude of a swarm: exponential base · case 02
Totals are many orders of magnitude too small.
Equivalent magnitude of a swarm: exponential base · case 03
Totals are many orders of magnitude too small.
Equivalent magnitude of a swarm: exponential base · case 04
Totals are many orders of magnitude too small.
Equivalent magnitude of a swarm: exponential base · case 05
Totals are many orders of magnitude too small.
Teleseismic body-wave magnitude: amplitude over period · case 01
Long-period readings inflate mb.
Teleseismic body-wave magnitude: amplitude over period · case 02
Long-period readings inflate mb.
Teleseismic body-wave magnitude: amplitude over period · case 03
Long-period readings inflate mb.
Teleseismic body-wave magnitude: amplitude over period · case 04
Long-period readings inflate mb.
Teleseismic body-wave magnitude: amplitude over period · case 05
Long-period readings inflate mb.
Teleseismic body-wave magnitude: calibration interpolation · case 01
Between table nodes the correction runs backward.
Teleseismic body-wave magnitude: calibration interpolation · case 02
Between table nodes the correction runs backward.
Teleseismic body-wave magnitude: calibration interpolation · case 03
Between table nodes the correction runs backward.
Teleseismic body-wave magnitude: calibration interpolation · case 04
Between table nodes the correction runs backward.
Teleseismic body-wave magnitude: calibration interpolation · case 05
Between table nodes the correction runs backward.
Teleseismic body-wave magnitude: distance validity range · case 01
Stations at exactly 100 degrees are rejected.
Teleseismic body-wave magnitude: distance validity range · case 02
Stations at exactly 100 degrees are rejected.
Teleseismic body-wave magnitude: distance validity range · case 03
Stations at exactly 100 degrees are rejected.
Teleseismic body-wave magnitude: distance validity range · case 04
Stations at exactly 100 degrees are rejected.
Teleseismic body-wave magnitude: distance validity range · case 05
Stations at exactly 100 degrees are rejected.
Teleseismic body-wave magnitude: short-period limit · case 01
3.0 s readings are discarded.
Teleseismic body-wave magnitude: short-period limit · case 02
3.0 s readings are discarded.
Teleseismic body-wave magnitude: short-period limit · case 03
3.0 s readings are discarded.
Teleseismic body-wave magnitude: short-period limit · case 04
3.0 s readings are discarded.
Teleseismic body-wave magnitude: short-period limit · case 05
3.0 s readings are discarded.
Prague surface-wave magnitude: period window · case 01
18 s Rayleigh readings are excluded.
Prague surface-wave magnitude: period window · case 02
18 s Rayleigh readings are excluded.
Prague surface-wave magnitude: period window · case 03
18 s Rayleigh readings are excluded.
Prague surface-wave magnitude: period window · case 04
18 s Rayleigh readings are excluded.
Prague surface-wave magnitude: period window · case 05
18 s Rayleigh readings are excluded.
Prague surface-wave magnitude: distance term log base · case 01
Ms is about 1.5 units too high.
Prague surface-wave magnitude: distance term log base · case 02
Ms is about 1.5 units too high.
Prague surface-wave magnitude: distance term log base · case 03
Ms is about 1.5 units too high.
Prague surface-wave magnitude: distance term log base · case 04
Ms is about 1.5 units too high.
Prague surface-wave magnitude: distance term log base · case 05
Ms is about 1.5 units too high.
Prague surface-wave magnitude: shallow depth limit · case 01
Events at exactly 60 km receive an Ms.
Prague surface-wave magnitude: shallow depth limit · case 02
Events at exactly 60 km receive an Ms.
Prague surface-wave magnitude: shallow depth limit · case 03
Events at exactly 60 km receive an Ms.
Prague surface-wave magnitude: shallow depth limit · case 04
Events at exactly 60 km receive an Ms.
Prague surface-wave magnitude: shallow depth limit · case 05
Events at exactly 60 km receive an Ms.
Prague surface-wave magnitude: station averaging · case 01
Event Ms follows the single loudest station.
Prague surface-wave magnitude: station averaging · case 02
Event Ms follows the single loudest station.
Prague surface-wave magnitude: station averaging · case 03
Event Ms follows the single loudest station.
Prague surface-wave magnitude: station averaging · case 04
Event Ms follows the single loudest station.
Prague surface-wave magnitude: station averaging · case 05
Event Ms follows the single loudest station.
Coda duration magnitude: duration origin · case 01
Records with long pre-event buffers yield inflated durations.
Coda duration magnitude: duration origin · case 02
Records with long pre-event buffers yield inflated durations.
Coda duration magnitude: duration origin · case 03
Records with long pre-event buffers yield inflated durations.
Coda duration magnitude: duration origin · case 04
Records with long pre-event buffers yield inflated durations.
Coda duration magnitude: duration origin · case 05
Records with long pre-event buffers yield inflated durations.
Coda duration magnitude: coda termination level · case 01
Coda end is delayed when the envelope settles exactly at twice the noise.
Coda duration magnitude: coda termination level · case 02
Coda end is delayed when the envelope settles exactly at twice the noise.
Coda duration magnitude: coda termination level · case 03
Coda end is delayed when the envelope settles exactly at twice the noise.
Coda duration magnitude: coda termination level · case 04
Coda end is delayed when the envelope settles exactly at twice the noise.
Coda duration magnitude: coda termination level · case 05
Coda end is delayed when the envelope settles exactly at twice the noise.
Coda duration magnitude: termination search start · case 01
A quiet sample right after the pick terminates the coda before the S wave peak.
Coda duration magnitude: termination search start · case 02
A quiet sample right after the pick terminates the coda before the S wave peak.
Coda duration magnitude: termination search start · case 03
A quiet sample right after the pick terminates the coda before the S wave peak.
Coda duration magnitude: termination search start · case 04
A quiet sample right after the pick terminates the coda before the S wave peak.
Coda duration magnitude: termination search start · case 05
A quiet sample right after the pick terminates the coda before the S wave peak.
Coda duration magnitude: duration log base · case 01
Md grows too fast with duration.
Coda duration magnitude: duration log base · case 02
Md grows too fast with duration.
Coda duration magnitude: duration log base · case 03
Md grows too fast with duration.
Coda duration magnitude: duration log base · case 04
Md grows too fast with duration.
Coda duration magnitude: duration log base · case 05
Md grows too fast with duration.
Aki maximum-likelihood b-value: completeness filter · case 01
Events exactly at the completeness magnitude are dropped and b is biased low.
Aki maximum-likelihood b-value: completeness filter · case 02
Events exactly at the completeness magnitude are dropped and b is biased low.
Aki maximum-likelihood b-value: completeness filter · case 03
Events exactly at the completeness magnitude are dropped and b is biased low.
Aki maximum-likelihood b-value: completeness filter · case 04
Events exactly at the completeness magnitude are dropped and b is biased low.
Aki maximum-likelihood b-value: completeness filter · case 05
Events exactly at the completeness magnitude are dropped and b is biased low.
Aki maximum-likelihood b-value: binning correction · case 01
b-values are systematically too high for binned catalogues.
Aki maximum-likelihood b-value: binning correction · case 02
b-values are systematically too high for binned catalogues.
Aki maximum-likelihood b-value: binning correction · case 03
b-values are systematically too high for binned catalogues.
Aki maximum-likelihood b-value: binning correction · case 04
b-values are systematically too high for binned catalogues.
Aki maximum-likelihood b-value: binning correction · case 05
b-values are systematically too high for binned catalogues.
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 ↗