Data Recovery Case File · Second Fixes & Trade Handoffs · Read the Output
Duplicates That Will Not Open Are the Signature of Carving Without Structures
His enquiry describes a result rather than a fault, and the result is diagnostic. A laptop that crashed and would not restart, where a previous provider "recovered a lot of the data, but many of the files are missing and there are multiple copies of many of the files, most of which don't open" — with a doctorate on the drive. That combination is not carelessness: it is what recovery looks like when the filesystem could not be used and content was located by pattern instead.
| Media | Laptop hard drive following a prior recovery attempt — output containing substantial gaps alongside multiple non-opening copies of many files; original drive retained by the owner |
| Reported situation | Laptop crashing and not restarting · prior recovery performed by another provider · substantial quantity of content returned · many files absent from the output · many files present in multiple copies · most duplicate copies not opening · original drive retained · academic work held on the drive |
| Fault class | Prior recovery performed by signature carving without filesystem structures — duplicates and gaps intrinsic to that method; structural reconstruction not yet attempted |
| Equipment used | Prior output analysed as evidence of method before the drive was re-examined · original drive imaged write-blocked under strict per-sector timeouts · filesystem structures interpreted with their duplicate copies before any carving · directory reconstruction attempted in preference to signature recovery · recovered content reconciled against the prior output to identify genuine gaps |
The decode: what the pattern reveals, and why a second attempt can do better
Why duplicates appear at all: the same content exists in several places. A drive holds the original file, thumbnails, application caches, temporary copies and previous versions, and each is a separate recoverable object.
Why a filesystem normally hides that: the directory describes only the current version. A recovery that can read the structures returns one copy of each file with its name and folder intact.
What happens when the structures cannot be read: the method changes. Content is located by recognising file openings across the whole surface, which finds every copy of everything with no way to tell which is current.
Why most of them do not open: the fragments are incomplete. A carve finds a beginning and estimates how far the file runs, and where content is not contiguous that estimate is wrong — producing a file with a valid opening and invalid continuation.
Why files are missing in the same output: the same limitation. Formats without a recognisable opening cannot be found by carving at all, and neither can anything whose first fragment is unreadable.
Why that adds up to a coherent picture: gaps and duplicates are two faces of one method. An output with both was carved rather than reconstructed, and the previous provider was doing the fallback rather than the primary work.
Why that leaves room to do better: the primary work may not have been attempted. Filesystems keep duplicate copies of their key records elsewhere on the volume, and reconstruction from those returns files with names, folders and correct lengths.
Why that matters especially for his material: a thesis is a small number of documents with a great deal of structure around them. Names, versions and folder arrangement are much of what makes the work usable, and carving destroys exactly that.
Why the prior output is worth keeping rather than discarding: it is a floor. Whatever it holds is already recovered, and comparing against it identifies what is genuinely still missing.
What the honest caveat is: if the structures were unreadable when they tried, they may still be. A second attempt is worth making and is not guaranteed to change the method available.
On the bench
The prior output was analysed as evidence of method before the drive was re-examined — duplicates arising because originals, thumbnails, caches, temporary copies and prior versions are separate recoverable objects that a readable directory would resolve to one current file each. Carving finds every copy and estimates extents, producing valid openings with invalid continuations, while formats lacking recognisable openings are not found at all. Directory reconstruction was attempted in preference to signature recovery.
The outcome
The prior output read as evidence of method, structures interpreted with their duplicate copies before any carving, and results reconciled against the earlier recovery. Free assessment, one fixed written figure including VAT; where a drive has to be opened, 50% of parts and labour is payable upfront with the balance only on success — otherwise no recovery, no fee. The decode: gaps and duplicates are two faces of one method. That output was carved rather than reconstructed — which means the approach that returns names and folders may not have been tried.
Reading the output of a recovery you already paid for
Keep it — whatever it holds is already recovered, and comparing against it identifies what's genuinely still missing. Then read the pattern: many copies of the same file, most of which don't open, alongside other files absent entirely, is the signature of content located by pattern across the surface rather than through the filesystem. Duplicates appear because thumbnails, caches and previous versions are separate objects a directory would have resolved. It's worth asking whether structural reconstruction was attempted at all.
Keep the output — call Cambridge Data Recovery on 01223 655015; prior results read as evidence of method, structures interpreted before any carving, findings reconciled against the earlier attempt.
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.