Data Recovery Case File · Trust, Practice & Honest Limits · The One Thing That Really Destroys
Overwriting Is the Only Common Action That Removes Content Rather Than Its Description
His enquiry states the worst of it without being asked, which makes an honest answer possible. A 128GB card "that has been formatted and written over, maybe twice", on which he wants an attempt made — with a straightforward question about process and cost. Almost every other fault leaves the content in place, and this is the exception: what has been written over is not recoverable by anyone, and the useful question is how much that is.
| Media | 128GB memory card formatted and subsequently reused, possibly across two cycles — original content partially overwritten; recovery attempt requested with realistic expectations |
| Reported situation | Memory card formatted following a loss of content · card returned to use afterwards · further content written, possibly across two separate periods · original material sought · owner requesting an attempt with realistic expectations · process and indicative cost queried |
| Fault class | Content partially overwritten by subsequent use — recoverable proportion determined by how much of the capacity later writing occupied and where it fell |
| Equipment used | Overwritten proportion established at assessment before any figure was quoted · card imaged write-blocked at genuine physical capacity · later content mapped and its extent measured against the card's capacity · original content carved by signature from regions later writing did not occupy · findings reported before recovery was scoped |
The decode: what overwriting actually removes, and how much is likely left
Why almost every other fault is recoverable and this one partly is not: the others remove descriptions. Deleting, formatting and corruption all affect the records that locate content, leaving the content itself in place.
What overwriting does differently: it replaces. New content written to a region physically replaces what was there, and nothing recovers what is no longer present.
Why that makes the question quantitative rather than binary: overwriting is partial. What matters is how much of the card later use occupied, not that it was used.
Why his estimate of twice is useful: it bounds the exposure. Two periods of use on a large card may still have filled only a fraction of it.
Why capacity works in his favour: the card is large. A hundred and twenty-eight gigabytes is a lot to fill, and casual use rarely reaches the end of it.
Where new content is written, which decides everything: from the start. A formatted card is filled from the beginning, so later use overwrites the earliest material first and the later material last.
What follows practically: whatever was at the end of the card is likeliest to survive. The final material written before the format is the most protected, which is the opposite of what people expect.
Why the format itself may have done more damage than the reuse: it depends on the type. Where a card is formatted in a device that issues an erase instruction, the memory may have been blanked rather than merely marked available.
Why that is established rather than assumed: the cell state is readable. Sampling the memory directly distinguishes blanked regions from populated ones before anything is quoted.
Why the honest order matters here more than usual: the assessment may be the answer. Where the card is largely blank or largely rewritten, that is worth knowing before any figure is agreed.
On the bench
The overwritten proportion was established at assessment before any figure was quoted — deletion, formatting and corruption affecting the records that locate content while leaving it in place, whereas overwriting physically replaces it. A formatted card fills from the beginning, so later use overwrites the earliest material first and the final material written before the format is the most protected. Cell state was sampled to distinguish blanked regions from populated ones.
The outcome
Overwritten proportion established before quoting, the card imaged at genuine capacity, and original content carved from unoccupied regions. Free assessment, one fixed written figure including VAT; on cards where content has been deleted or overwritten, the figure is payable upfront. The decode: overwriting is the one action that genuinely removes. But a large card fills from the beginning, so what survives best is whatever was written last before the format — which is usually not what people assume.
A card that has been reused since the loss
Stop using it now, and say plainly how much use it has had — it makes an honest answer possible. Overwriting is the one common action that actually removes content rather than its description, so the question is quantitative: how much of the card later use occupied. A large card fills from the beginning, which means the earliest material is overwritten first and whatever was written last before the format survives best. Expect the assessment itself to be the answer.
Stop using it — call Cambridge Data Recovery on 01223 655015; overwritten proportion established before any figure, cell state sampled directly, findings reported before recovery is scoped.
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.