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
Partial subtitle resync from a point: negative start clamp · case 01
Cues that straddle zero after an early shift disappear entirely.
Partial subtitle resync from a point: negative start clamp · case 02
Cues that straddle zero after an early shift disappear entirely.
Partial subtitle resync from a point: negative start clamp · case 03
Cues that straddle zero after an early shift disappear entirely.
Partial subtitle resync from a point: negative start clamp · case 04
Cues that straddle zero after an early shift disappear entirely.
Partial subtitle resync from a point: negative start clamp · case 05
Cues that straddle zero after an early shift disappear entirely.
Partial subtitle resync from a point: resort after partial shift · case 01
After a negative partial resync, cue numbers no longer follow display order.
Partial subtitle resync from a point: resort after partial shift · case 02
After a negative partial resync, cue numbers no longer follow display order.
Partial subtitle resync from a point: resort after partial shift · case 03
After a negative partial resync, cue numbers no longer follow display order.
Partial subtitle resync from a point: resort after partial shift · case 04
After a negative partial resync, cue numbers no longer follow display order.
Partial subtitle resync from a point: resort after partial shift · case 05
After a negative partial resync, cue numbers no longer follow display order.
Partial subtitle resync from a point: end time shift · case 01
Shifted cues keep their old end time and become too long or inverted.
Partial subtitle resync from a point: end time shift · case 02
Shifted cues keep their old end time and become too long or inverted.
Partial subtitle resync from a point: end time shift · case 03
Shifted cues keep their old end time and become too long or inverted.
Partial subtitle resync from a point: end time shift · case 04
Shifted cues keep their old end time and become too long or inverted.
Partial subtitle resync from a point: end time shift · case 05
Shifted cues keep their old end time and become too long or inverted.
Subtitle frame-rate retiming: scale direction · case 01
After a PAL speed-up conversion subtitles drift later and later behind the dialogue.
Subtitle frame-rate retiming: scale direction · case 02
After a PAL speed-up conversion subtitles drift later and later behind the dialogue.
Subtitle frame-rate retiming: scale direction · case 03
After a PAL speed-up conversion subtitles drift later and later behind the dialogue.
Subtitle frame-rate retiming: scale direction · case 04
After a PAL speed-up conversion subtitles drift later and later behind the dialogue.
Subtitle frame-rate retiming: scale direction · case 05
After a PAL speed-up conversion subtitles drift later and later behind the dialogue.
Subtitle frame-rate retiming: rational rate parsing · case 01
NTSC-rate sources written as 30000/1001 retime as if they were exactly 29 fps.
Subtitle frame-rate retiming: rational rate parsing · case 02
NTSC-rate sources written as 30000/1001 retime as if they were exactly 29 fps.
Subtitle frame-rate retiming: rational rate parsing · case 03
NTSC-rate sources written as 30000/1001 retime as if they were exactly 29 fps.
Subtitle frame-rate retiming: rational rate parsing · case 04
NTSC-rate sources written as 30000/1001 retime as if they were exactly 29 fps.
Subtitle frame-rate retiming: rational rate parsing · case 05
NTSC-rate sources written as 30000/1001 retime as if they were exactly 29 fps.
Subtitle frame-rate retiming: ntsc alias rate · case 01
Files labelled 23.976 accumulate a small drift against the real 24000/1001 video.
Subtitle frame-rate retiming: ntsc alias rate · case 02
Files labelled 23.976 accumulate a small drift against the real 24000/1001 video.
Subtitle frame-rate retiming: ntsc alias rate · case 03
Files labelled 23.976 accumulate a small drift against the real 24000/1001 video.
Subtitle frame-rate retiming: ntsc alias rate · case 04
Files labelled 23.976 accumulate a small drift against the real 24000/1001 video.
Subtitle frame-rate retiming: ntsc alias rate · case 05
Files labelled 23.976 accumulate a small drift against the real 24000/1001 video.
Subtitle frame-rate retiming: start rounding · case 01
Cue starts land one millisecond early whenever the scaled time has a fractional half or more.
Subtitle frame-rate retiming: start rounding · case 02
Cue starts land one millisecond early whenever the scaled time has a fractional half or more.
Subtitle frame-rate retiming: start rounding · case 03
Cue starts land one millisecond early whenever the scaled time has a fractional half or more.
Subtitle frame-rate retiming: start rounding · case 04
Cue starts land one millisecond early whenever the scaled time has a fractional half or more.
Subtitle frame-rate retiming: start rounding · case 05
Cue starts land one millisecond early whenever the scaled time has a fractional half or more.
Subtitle frame-rate retiming: independent end rounding · case 01
Cue ends are off by a millisecond relative to the independently scaled timeline.
Subtitle frame-rate retiming: independent end rounding · case 02
Cue ends are off by a millisecond relative to the independently scaled timeline.
Subtitle frame-rate retiming: independent end rounding · case 03
Cue ends are off by a millisecond relative to the independently scaled timeline.
Subtitle frame-rate retiming: independent end rounding · case 04
Cue ends are off by a millisecond relative to the independently scaled timeline.
Subtitle frame-rate retiming: independent end rounding · case 05
Cue ends are off by a millisecond relative to the independently scaled timeline.
Cue snapping to video frame grid: start frame floor · case 01
Cues appear a frame after the dialogue begins.
Cue snapping to video frame grid: start frame floor · case 02
Cues appear a frame after the dialogue begins.
Cue snapping to video frame grid: start frame floor · case 03
Cues appear a frame after the dialogue begins.
Cue snapping to video frame grid: start frame floor · case 04
Cues appear a frame after the dialogue begins.
Cue snapping to video frame grid: start frame floor · case 05
Cues appear a frame after the dialogue begins.
Cue snapping to video frame grid: end frame ceiling · case 01
Cues vanish one frame before the spoken line ends.
Cue snapping to video frame grid: end frame ceiling · case 02
Cues vanish one frame before the spoken line ends.
Cue snapping to video frame grid: end frame ceiling · case 03
Cues vanish one frame before the spoken line ends.
Cue snapping to video frame grid: end frame ceiling · case 04
Cues vanish one frame before the spoken line ends.
Cue snapping to video frame grid: end frame ceiling · case 05
Cues vanish one frame before the spoken line ends.
Cue snapping to video frame grid: overlap push test · case 01
Consecutive snapped cues share frames and flicker on screen together.
Cue snapping to video frame grid: overlap push test · case 02
Consecutive snapped cues share frames and flicker on screen together.
Cue snapping to video frame grid: overlap push test · case 03
Consecutive snapped cues share frames and flicker on screen together.
Cue snapping to video frame grid: overlap push test · case 04
Consecutive snapped cues share frames and flicker on screen together.
Cue snapping to video frame grid: overlap push test · case 05
Consecutive snapped cues share frames and flicker on screen together.
Cue snapping to video frame grid: previous end tracking · case 01
Snapped cues overlap by several frames after a long cue.
Cue snapping to video frame grid: previous end tracking · case 02
Snapped cues overlap by several frames after a long cue.
Cue snapping to video frame grid: previous end tracking · case 03
Snapped cues overlap by several frames after a long cue.
Cue snapping to video frame grid: previous end tracking · case 04
Snapped cues overlap by several frames after a long cue.
Cue snapping to video frame grid: previous end tracking · case 05
Snapped cues overlap by several frames after a long cue.
Cue snapping to video frame grid: frame to millisecond conversion · case 01
Millisecond boundaries of NTSC-rate cues are about 0.1 percent early.
Cue snapping to video frame grid: frame to millisecond conversion · case 02
Millisecond boundaries of NTSC-rate cues are about 0.1 percent early.
Cue snapping to video frame grid: frame to millisecond conversion · case 03
Millisecond boundaries of NTSC-rate cues are about 0.1 percent early.
Cue snapping to video frame grid: frame to millisecond conversion · case 04
Millisecond boundaries of NTSC-rate cues are about 0.1 percent early.
Cue snapping to video frame grid: frame to millisecond conversion · case 05
Millisecond boundaries of NTSC-rate cues are about 0.1 percent early.
Minimum gap and chaining between cues: trim trigger · case 01
Cues with a tiny positive gap blink the screen for a couple of frames between lines.
Minimum gap and chaining between cues: trim trigger · case 02
Cues with a tiny positive gap blink the screen for a couple of frames between lines.
Minimum gap and chaining between cues: trim trigger · case 03
Cues with a tiny positive gap blink the screen for a couple of frames between lines.
Minimum gap and chaining between cues: trim trigger · case 04
Cues with a tiny positive gap blink the screen for a couple of frames between lines.
Minimum gap and chaining between cues: trim trigger · case 05
Cues with a tiny positive gap blink the screen for a couple of frames between lines.
Minimum gap and chaining between cues: trim floor · case 01
A short cue followed immediately by another ends before it starts.
Minimum gap and chaining between cues: trim floor · case 02
A short cue followed immediately by another ends before it starts.
Minimum gap and chaining between cues: trim floor · case 03
A short cue followed immediately by another ends before it starts.
Minimum gap and chaining between cues: trim floor · case 04
A short cue followed immediately by another ends before it starts.
Minimum gap and chaining between cues: trim floor · case 05
A short cue followed immediately by another ends before it starts.
Minimum gap and chaining between cues: chain window · case 01
Gaps of exactly the chaining threshold are closed although the contract leaves them open.
Minimum gap and chaining between cues: chain window · case 02
Gaps of exactly the chaining threshold are closed although the contract leaves them open.
Minimum gap and chaining between cues: chain window · case 03
Gaps of exactly the chaining threshold are closed although the contract leaves them open.
Minimum gap and chaining between cues: chain window · case 04
Gaps of exactly the chaining threshold are closed although the contract leaves them open.
Minimum gap and chaining between cues: chain window · case 05
Gaps of exactly the chaining threshold are closed although the contract leaves them open.
Minimum gap and chaining between cues: chain target · case 01
Chained cues touch the next cue with no separating gap.
Minimum gap and chaining between cues: chain target · case 02
Chained cues touch the next cue with no separating gap.
Minimum gap and chaining between cues: chain target · case 03
Chained cues touch the next cue with no separating gap.
Minimum gap and chaining between cues: chain target · case 04
Chained cues touch the next cue with no separating gap.
Minimum gap and chaining between cues: chain target · case 05
Chained cues touch the next cue with no separating gap.
Minimum gap and chaining between cues: gap measurement · case 01
Almost no cue is ever trimmed or chained because gaps look enormous.
Minimum gap and chaining between cues: gap measurement · case 02
Almost no cue is ever trimmed or chained because gaps look enormous.
Minimum gap and chaining between cues: gap measurement · case 03
Almost no cue is ever trimmed or chained because gaps look enormous.
Minimum gap and chaining between cues: gap measurement · case 04
Almost no cue is ever trimmed or chained because gaps look enormous.
Minimum gap and chaining between cues: gap measurement · case 05
Almost no cue is ever trimmed or chained because gaps look enormous.
Minimum cue duration extension: extension base · case 01
Short cues become much longer than the minimum duration.
Minimum cue duration extension: extension base · case 02
Short cues become much longer than the minimum duration.
Minimum cue duration extension: extension base · case 03
Short cues become much longer than the minimum duration.
Minimum cue duration extension: extension base · case 04
Short cues become much longer than the minimum duration.
Minimum cue duration extension: extension base · case 05
Short cues become much longer than the minimum duration.
Minimum cue duration extension: neighbour gap limit · case 01
Extended cues butt directly against the following cue.
Minimum cue duration extension: neighbour gap limit · case 02
Extended cues butt directly against the following cue.
Minimum cue duration extension: neighbour gap limit · case 03
Extended cues butt directly against the following cue.
Minimum cue duration extension: neighbour gap limit · case 04
Extended cues butt directly against the following cue.
Minimum cue duration extension: neighbour gap limit · case 05
Extended cues butt directly against the following cue.
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 ↗