Data Recovery Case File · Cameras, Drones & Cards · A Split That Follows the Recording
Files Failing in the Order They Were Written Points at a Region, Not at the Files
His enquiry contains a pattern and a refreshingly narrow objective. A card from an action camera where "half of the videos are okay and playable, however the second half are unplayable" and online services could not help — and he wants one file, around 2.2 gigabytes and twelve minutes long. The split is the diagnosis: cards fill in order, so a failure affecting the later half is a failure affecting the later part of the card.
| Media | Memory card from an action camera — earlier recordings playing normally, later recordings not playable; one specific video file of approximately 2.2GB sought |
| Reported situation | Memory card used continuously in an action camera · earlier half of recorded videos playing correctly · later half not playable · consumer recovery services attempted without success · one specific file identified as the objective · file size and duration supplied by the owner |
| Fault class | Regional degradation affecting later-written areas — earlier content unaffected; target file recoverable in proportion to how much of its extent falls within the degraded region |
| Equipment used | Failure pattern mapped against write order rather than against file type · card read on fixed contact equipment independent of host enumeration · imaged with per-region read verification and unstable areas re-read across repeated passes · target file located by size and duration and its extent mapped against the degraded region · stream repaired and validated by playback |
The decode: why the split is positional, and what it means for his one file
Why a card fills in order: allocation is sequential on fresh media. A camera writes each recording after the last, so file order and physical order match closely.
Why that turns his observation into a map: the boundary is a place. Files failing from a certain point onward means the card degrades from a certain position onward, rather than certain files being individually unlucky.
Why that rules out several explanations at once: it is not the camera and not the format. A recorder writing badly would produce scattered failures, and a format problem would affect the whole card.
Why later regions fail first on a card used this way: they are reached last and least. Regions written toward the end of a card's capacity see fewer cycles but also receive less of the controller's attention, and a manufacturing weakness there goes unnoticed until the card is filled.
Why consumer services could not help: they ask the card for the data. A tool that requests a region the card cannot return correctly receives what the card hands over, which is wrong.
What is done instead: repeated reading and comparison. A marginal region returns differently on different attempts, and many passes reconstruct a correct result for a good proportion of it.
Now his single-file objective, which changes the work usefully: effort concentrates. Knowing which file matters means the passes are spent on the regions holding it rather than spread across the card.
Why the size and duration he supplied are genuinely useful: they identify the file without a directory. A twelve-minute recording of a known size is locatable by its own structure even where records are damaged.
What determines the outcome for that file specifically: where it sits relative to the boundary. A file beginning before the degraded region and running into it recovers partly and plays up to the damage.
Why video is forgiving in a way most formats are not: it is a continuous stream. A damaged section costs an interval rather than the whole file, and playback frequently resumes afterwards.
On the bench
The failure pattern was mapped against write order rather than against file type — cards allocating sequentially so file order and physical order correspond closely, which makes failure from a given file onward a failure from a given position onward, and excludes both recorder faults, which scatter, and format faults, which affect everything. The target file was located by size and duration and its extent mapped against the degraded region before effort was concentrated there.
The outcome
The pattern mapped against write order, unstable regions re-read across repeated passes, and effort concentrated on the identified file. Free assessment, one fixed written figure including VAT; on cards where content has been deleted or overwritten, the figure is payable upfront. The decode: cards fill in order, so half your videos failing is half your card failing — and telling us which single file matters means the passes are spent where it actually sits.
When later files fail and earlier ones are fine
Say where the boundary falls, because it's a map rather than an anecdote: cards allocate sequentially, so failure from a certain point in the recording order onward means degradation from a certain point on the card. That rules out the camera, which would scatter failures, and the format, which would affect everything. Name the one file you actually need too — it concentrates the repeated reading on the regions holding it, and a size and duration identify a file even without its directory record.
Say where the split falls — call Cambridge Data Recovery on 01223 655015; the pattern mapped against write order, regions re-read across repeated passes, effort concentrated on the file you name.
Request a quote online →
Our case files are drawn from genuine enquiries received by our laboratory over the past ten years, anonymised to protect client confidentiality. Each one describes the diagnostic and recovery procedure our engineers apply to that fault, using the equipment listed.