Memory cards · job record · PLY-2025-0907
Files the Right Size, and Nothing That Plays.
The card had turned against him overnight. A videographer in Camborne sent it in after a marquee wedding on a farm towards Carn Brea, with a plain description: the evening's last three files simply will not import
. What he added next framed the job — the card still lists them at full size, so the video is in there somewhere
— and the couple's flight was less than two weeks away.
Sounds like yours? Talk it through.
0800 6890668
Why it happens.
His reading of it was right. Recording happens in two halves. Frame after frame goes onto the card as the shot is made. Written last of all, at the moment a file is closed in good order, is the index: a list giving the start point of every frame and its duration. Yank the card, drain the battery or fill the card up, and the frames remain while the table never gets written. Nothing but that table tells an editor how to move through a clip. Which is why a file crammed with video and carrying no map looks identical on screen to one holding nothing.
The kit this job called for.
How the job runs →| The kit | What it did here | Why we keep it |
|---|---|---|
| Klennet Carver | Assembled the surviving frames into files his editing suite would accept | It knows how a video file is built inside; blunter tools return fragments that play nowhere |
| PC-3000 Flash | Made the complete sector-by-sector copy of the card ahead of everything else | Bypasses the controller and reads the raw NAND, against a maker-ID library we keep up to date |
| UFS Explorer Professional Recovery | Recovered the sound clips and the folder layout out of that copy | Mounts what other software will not: APFS, ReFS, XFS, ZFS and Btrfs volumes |
How it ran.
Copy first, diagnose afterwards
Not a word was said about the fault until a copy existed. A date already promised to somebody else will change the pace a job is run at. What it does not buy anybody is a free hand with the only original in the room. Everything from that point was done against the image. The card itself went into an antistatic bag and was set aside, should a second read ever be needed.
Secure the clips that still open before anything else
Almost all of the evening lifted straight off the copy and went somewhere safe before anything harder was tried. The three awkward files were opened and read through after that, and not before. He had read it right: frame data in all three, an index half-written in two, and none whatever in the third. The order of work followed from that.
Borrow the pattern from the clips that came through
Video seldom goes down in one continuous run, and the scattering is where one-click software becomes useless — what it returns will play nowhere. A pattern came from undamaged clips shot earlier that evening on the same body at identical settings. Every broken file was then reconstructed around whatever frames remained in it, and tested the one way worth anything: played from first frame to last.
How it ended up.
Two of the three came back entire, and both were watched right through on the bench before they went out. The third plays cleanly to within a few seconds of the finish; those seconds were still sitting in the camera buffer when the card came out. The edit reached the couple with the flight still ahead of them.
Pages people open after this.
More cases filed under cards.
Seeing the same thing?
Leave it switched off and post it in. Nothing happens until the diagnosis, which lists what can still be read and what cannot.