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
EDL speed-change record out: midnight source span · case 01
Source clips that cross midnight yield negative or near-day record durations.
EDL speed-change record out: midnight source span · case 02
Source clips that cross midnight yield negative or near-day record durations.
EDL speed-change record out: midnight source span · case 03
Source clips that cross midnight yield negative or near-day record durations.
EDL speed-change record out: midnight source span · case 04
Source clips that cross midnight yield negative or near-day record durations.
EDL speed-change record out: midnight source span · case 05
Source clips that cross midnight yield negative or near-day record durations.
EDL speed-change record out: record label wrap · case 01
Record out labels past midnight show hours of 24 or more.
EDL speed-change record out: record label wrap · case 02
Record out labels past midnight show hours of 24 or more.
EDL speed-change record out: record label wrap · case 03
Record out labels past midnight show hours of 24 or more.
EDL speed-change record out: record label wrap · case 04
Record out labels past midnight show hours of 24 or more.
EDL speed-change record out: record label wrap · case 05
Record out labels past midnight show hours of 24 or more.
EDL speed-change record out: record out exclusivity · case 01
Record out labels are one frame early, colliding with the next event.
EDL speed-change record out: record out exclusivity · case 02
Record out labels are one frame early, colliding with the next event.
EDL speed-change record out: record out exclusivity · case 03
Record out labels are one frame early, colliding with the next event.
EDL speed-change record out: record out exclusivity · case 04
Record out labels are one frame early, colliding with the next event.
EDL speed-change record out: record out exclusivity · case 05
Record out labels are one frame early, colliding with the next event.
Film feet and frames conversion: negative sign emission · case 01
Negative footage counts print as positive footage.
Film feet and frames conversion: negative sign emission · case 02
Negative footage counts print as positive footage.
Film feet and frames conversion: negative sign emission · case 03
Negative footage counts print as positive footage.
Film feet and frames conversion: negative sign emission · case 04
Negative footage counts print as positive footage.
Film feet and frames conversion: negative sign emission · case 05
Negative footage counts print as positive footage.
Film feet and frames conversion: frame field bound · case 01
A frame field equal to the frames per foot is accepted.
Film feet and frames conversion: frame field bound · case 02
A frame field equal to the frames per foot is accepted.
Film feet and frames conversion: frame field bound · case 03
A frame field equal to the frames per foot is accepted.
Film feet and frames conversion: frame field bound · case 04
A frame field equal to the frames per foot is accepted.
Film feet and frames conversion: frame field bound · case 05
A frame field equal to the frames per foot is accepted.
Film feet and frames conversion: negative zero feet · case 01
Negative footage within the first foot, such as -0+05, is read as positive.
Film feet and frames conversion: negative zero feet · case 02
Negative footage within the first foot, such as -0+05, is read as positive.
Film feet and frames conversion: negative zero feet · case 03
Negative footage within the first foot, such as -0+05, is read as positive.
Film feet and frames conversion: negative zero feet · case 04
Negative footage within the first foot, such as -0+05, is read as positive.
Film feet and frames conversion: negative zero feet · case 05
Negative footage within the first foot, such as -0+05, is read as positive.
Film feet and frames conversion: magnitude composition · case 01
Negative footage with frames converts to a count several frames too small.
Film feet and frames conversion: magnitude composition · case 02
Negative footage with frames converts to a count several frames too small.
Film feet and frames conversion: magnitude composition · case 03
Negative footage with frames converts to a count several frames too small.
Film feet and frames conversion: magnitude composition · case 04
Negative footage with frames converts to a count several frames too small.
Film feet and frames conversion: magnitude composition · case 05
Negative footage with frames converts to a count several frames too small.
Film feet and frames conversion: foot divisor · case 01
Converted 16mm footage shows three times too many feet.
Film feet and frames conversion: foot divisor · case 02
Converted 16mm footage shows three times too many feet.
Film feet and frames conversion: foot divisor · case 03
Converted 16mm footage shows three times too many feet.
Film feet and frames conversion: foot divisor · case 04
Converted 16mm footage shows three times too many feet.
Film feet and frames conversion: foot divisor · case 05
Converted 16mm footage shows three times too many feet.
2:3 pulldown film frame placement: cadence group size · case 01
Film frames after the first cadence group map to video frames far too early.
2:3 pulldown film frame placement: cadence group size · case 02
Film frames after the first cadence group map to video frames far too early.
2:3 pulldown film frame placement: cadence group size · case 03
Film frames after the first cadence group map to video frames far too early.
2:3 pulldown film frame placement: cadence group size · case 04
Film frames after the first cadence group map to video frames far too early.
2:3 pulldown film frame placement: cadence group size · case 05
Film frames after the first cadence group map to video frames far too early.
2:3 pulldown film frame placement: field offset table · case 01
B-frames are placed as if every film frame had two fields.
2:3 pulldown film frame placement: field offset table · case 02
B-frames are placed as if every film frame had two fields.
2:3 pulldown film frame placement: field offset table · case 03
B-frames are placed as if every film frame had two fields.
2:3 pulldown film frame placement: field offset table · case 04
B-frames are placed as if every film frame had two fields.
2:3 pulldown film frame placement: field offset table · case 05
B-frames are placed as if every film frame had two fields.
2:3 pulldown film frame placement: phase alignment · case 01
With a non-zero cadence phase the first film frame does not start at field 0.
2:3 pulldown film frame placement: phase alignment · case 02
With a non-zero cadence phase the first film frame does not start at field 0.
2:3 pulldown film frame placement: phase alignment · case 03
With a non-zero cadence phase the first film frame does not start at field 0.
2:3 pulldown film frame placement: phase alignment · case 04
With a non-zero cadence phase the first film frame does not start at field 0.
2:3 pulldown film frame placement: phase alignment · case 05
With a non-zero cadence phase the first film frame does not start at field 0.
2:3 pulldown film frame placement: field count lookup · case 01
Field counts are wrong whenever the cadence phase is not zero.
2:3 pulldown film frame placement: field count lookup · case 02
Field counts are wrong whenever the cadence phase is not zero.
2:3 pulldown film frame placement: field count lookup · case 03
Field counts are wrong whenever the cadence phase is not zero.
2:3 pulldown film frame placement: field count lookup · case 04
Field counts are wrong whenever the cadence phase is not zero.
2:3 pulldown film frame placement: field count lookup · case 05
Field counts are wrong whenever the cadence phase is not zero.
2:3 pulldown film frame placement: field parity · case 01
Frames starting on the second field of a video frame are reported as upper field first.
2:3 pulldown film frame placement: field parity · case 02
Frames starting on the second field of a video frame are reported as upper field first.
2:3 pulldown film frame placement: field parity · case 03
Frames starting on the second field of a video frame are reported as upper field first.
2:3 pulldown film frame placement: field parity · case 04
Frames starting on the second field of a video frame are reported as upper field first.
2:3 pulldown film frame placement: field parity · case 05
Frames starting on the second field of a video frame are reported as upper field first.
Reverse telecine field attribution: field to frame table · case 01
Mixed frames are reported at the wrong video frames.
Reverse telecine field attribution: field to frame table · case 02
Mixed frames are reported at the wrong video frames.
Reverse telecine field attribution: field to frame table · case 03
Mixed frames are reported at the wrong video frames.
Reverse telecine field attribution: field to frame table · case 04
Mixed frames are reported at the wrong video frames.
Reverse telecine field attribution: field to frame table · case 05
Mixed frames are reported at the wrong video frames.
Reverse telecine field attribution: second field index · case 01
The second field is taken from the previous video frame.
Reverse telecine field attribution: second field index · case 02
The second field is taken from the previous video frame.
Reverse telecine field attribution: second field index · case 03
The second field is taken from the previous video frame.
Reverse telecine field attribution: second field index · case 04
The second field is taken from the previous video frame.
Reverse telecine field attribution: second field index · case 05
The second field is taken from the previous video frame.
Reverse telecine field attribution: field order swap · case 01
Lower-field-first material reports the same film frame for both fields.
Reverse telecine field attribution: field order swap · case 02
Lower-field-first material reports the same film frame for both fields.
Reverse telecine field attribution: field order swap · case 03
Lower-field-first material reports the same film frame for both fields.
Reverse telecine field attribution: field order swap · case 04
Lower-field-first material reports the same film frame for both fields.
Reverse telecine field attribution: field order swap · case 05
Lower-field-first material reports the same film frame for both fields.
Reverse telecine field attribution: cadence base offset · case 01
With a non-zero phase, fields are attributed as if the cadence started at A.
Reverse telecine field attribution: cadence base offset · case 02
With a non-zero phase, fields are attributed as if the cadence started at A.
Reverse telecine field attribution: cadence base offset · case 03
With a non-zero phase, fields are attributed as if the cadence started at A.
Reverse telecine field attribution: cadence base offset · case 04
With a non-zero phase, fields are attributed as if the cadence started at A.
Reverse telecine field attribution: cadence base offset · case 05
With a non-zero phase, fields are attributed as if the cadence started at A.
Reverse telecine field attribution: phase frame rebasing · case 01
Film frame numbers are too high by the cadence phase.
Reverse telecine field attribution: phase frame rebasing · case 02
Film frame numbers are too high by the cadence phase.
Reverse telecine field attribution: phase frame rebasing · case 03
Film frame numbers are too high by the cadence phase.
Reverse telecine field attribution: phase frame rebasing · case 04
Film frame numbers are too high by the cadence phase.
Reverse telecine field attribution: phase frame rebasing · case 05
Film frame numbers are too high by the cadence phase.
Non-drop label to wall clock at pulled-down rates: pulldown denominator · case 01
An hour of 29.97 labels maps to exactly one hour of wall clock, 3.6 seconds short.
Non-drop label to wall clock at pulled-down rates: pulldown denominator · case 02
An hour of 29.97 labels maps to exactly one hour of wall clock, 3.6 seconds short.
Non-drop label to wall clock at pulled-down rates: pulldown denominator · case 03
An hour of 29.97 labels maps to exactly one hour of wall clock, 3.6 seconds short.
Non-drop label to wall clock at pulled-down rates: pulldown denominator · case 04
An hour of 29.97 labels maps to exactly one hour of wall clock, 3.6 seconds short.
Non-drop label to wall clock at pulled-down rates: pulldown denominator · case 05
An hour of 29.97 labels maps to exactly one hour of wall clock, 3.6 seconds short.
Non-drop label to wall clock at pulled-down rates: label to ms rounding · case 01
Elapsed times are a millisecond early for many frames.
Non-drop label to wall clock at pulled-down rates: label to ms rounding · case 02
Elapsed times are a millisecond early for many frames.
Non-drop label to wall clock at pulled-down rates: label to ms rounding · case 03
Elapsed times are a millisecond early for many frames.
Non-drop label to wall clock at pulled-down rates: label to ms rounding · case 04
Elapsed times are a millisecond early for many frames.
Non-drop label to wall clock at pulled-down rates: label to ms rounding · case 05
Elapsed times are a millisecond early for many frames.
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 ↗