A photographer shot over a card still holding a commercial job. Deleting is recoverable and formatting usually is; this is neither — and what survived came down to which blocks the card had not yet got round to erasing.
A professional photographer needed footage from a commercial shoot that had been on a Samsung EVO Select microSD card. New material had since been written to the same card. Consumer recovery software found nothing usable, and a client presentation was approaching.
The card itself was electrically perfect: no controller fault, no bad blocks, no physical damage. That is the awkward part of these cases — nothing is broken, and the loss in the affected areas is nonetheless total.
Worth setting out properly, because the mechanism is not what most people picture and it explains exactly why 65% survived rather than none or all.
On a hard disk, overwriting is literal: the head magnetises the same physical location differently and the previous state is gone.
Flash cannot do that. A page of flash can be programmed once, and to write it again the whole block containing it — typically hundreds of pages — must first be erased. So when a camera writes a new file to an address that already held one, the controller does not modify the old location at all. It programs the new data to a fresh page elsewhere, updates its translation table to point the address at the new page, and marks the old page stale.
The stale page still holds the original data. It stays that way until garbage collection reclaims the block it sits in — consolidating whatever remains valid, then erasing the whole block to make it writable again.
Which produces the pattern seen here. Where garbage collection had already reclaimed the blocks, the original footage was genuinely erased: the cells returned to their blank state, and nothing at any price retrieves that. Where it had not yet got round to them, the earlier shoot was still sitting there, invalidated but physically intact and readable at chip level.
So the boundary was not neat, and it did not correspond to anything the photographer could have predicted. It corresponded to which blocks the controller happened to have reclaimed.
None of the above is a loophole. Two things follow from it and both point the same way.
Garbage collection is continuous and runs whenever the card has power and spare capacity. Every additional minute the card spends in a camera or a reader gives it further opportunity to reclaim exactly the blocks holding what you want. The window closes on its own, without anyone doing anything.
And once a block is erased there is no residual signal to work from. This is a genuine difference from magnetic media, where the folklore about recovering overwritten data at least had a physical premise, however impractical. On flash the cells are returned to a blank state, and blank is blank.
We were straight with the photographer at the diagnostic: a full recovery was not available, and the realistic question was how much of the shoot fell in blocks that had not yet been reclaimed. That is a better conversation to have before an invoice than after one.
The card was imaged read-only behind a write blocker before anything else, and all work ran against the copy. Powering the card as little as possible is not a formality here — it is the difference between the material still being there and not.
The file system could not be trusted to describe what remained, so the image was carved by signature: scanned for the byte patterns that open video containers and camera RAW files, with each file reconstructed outward from its header.
Video needed an extra step, and it is the reason partial clips were returnable at all. Many cameras write the video data first and the index describing it last, at the end of the file. That index is what tells a player which frames exist and where they start. Lose it — because the tail of the clip fell in a reclaimed block — and the player sees a file full of perfectly good compressed frames and refuses to open it, because it has no map.
So the surviving frame data was extracted and remuxed into a new container with a freshly built index. The photographer got the first portion of a take as a playable file, instead of a large unopenable one.
Fragments too short to constitute usable footage were logged rather than returned as clutter.
Everything recovered was played end to end rather than counted. Partial clips were reported explicitly as partial with their durations, so the photographer could see what was usable before deciding anything. Embedded timestamps survived, which allowed the material to be placed back in shooting order.
About 65% of the shoot came back — the material that fell outside the reclaimed blocks, plus intact RAW stills. Enough to salvage the presentation, and considerably less than everything.
We publish this one because a page of 97s and 100s tells you nothing about a laboratory. Some jobs come back partially and the reason is usually physics rather than effort.
The practical instruction is simple and it is about the next minute rather than the last one: if a card holds something you need, take it out and stop using it. Not at the end of the shoot — now. Cards and USB media are from £250 +VAT after a free 48-hour diagnostic.
Post it or bring it in. Either way the diagnostic costs nothing and commits you to nothing, and you get one figure in writing before a single screw is turned.
Nothing can begin until the device is on the bench, which makes this the only step that needs anything from you. Pack it properly, put your details in with it, and send it over. What follows is a free diagnostic and a written figure, in that order.
Posting it? Use something tracked and insured — whatever is on the drive is worth considerably more than the postage. Bringing it in? Weekdays, 9am to 5:30pm, and it still wants packing as above for the journey.
Tell us what the device is, what it is doing, and anything already tried — an engineer will come back with a realistic view of the odds and a price band, before you commit to posting anything.
We’ll get back to you soon. Anything pressing, call 0161 871 0788.
Not the parts the card has already reclaimed. Flash cannot overwrite in place: the controller writes new data to a fresh page, repoints its table and marks the old page stale, and the original survives there until garbage collection erases the whole block containing it. Once that erase has run the cells are blank, and nothing recovers blank.
Deleting removes the pointer and leaves the data sitting there; a quick format rewrites the index and does much the same. Both are routinely recoverable. Shooting over a card is different in kind — it marks the old pages stale and starts a clock the card runs by itself, in the background, whenever it has power.
Eject it and leave it out of everything, now rather than at the end of the day. This is not the usual advice about avoiding further writes — a card reclaims stale blocks on its own schedule, so simply leaving it powered narrows what is left even if nobody touches it.
Often, in part — and it needs a step most tools skip. Many cameras write the frame data first and the index describing it last, so a clip whose tail was reclaimed is full of perfectly good compressed frames that no player will open, because the map is missing. Rebuilding an index around what survived turns that into a playable opening section rather than an unopenable file.
Cards and USB media are from £250 plus VAT, fixed in a written quote after a free 48-hour diagnostic. Where the diagnostic shows the material was overwritten, we say so before you commit rather than after.
Kick off with an instant online quote, or ring us and talk it through first. Either way you’ll know a clear, fixed price before any work starts.