A match is strong evidence of identical bytes
SHA-256 produces a 256-bit digest from file content. It is deterministic rather than statistically accurate: the same bytes produce the same digest. For a particular different input, the idealised chance of randomly producing the same digest is 1 in 2^256—about 1 in 1.16 × 10^77.
That does not make every verification process infallible. Results still depend on using the intended trusted reference, selecting the correct archive scope and successfully reading the files. A hash comparison establishes content agreement with the digest; it does not by itself establish the identity of the person who supplied the digest.
Algorithm versus manifest format
SHA-256 is the calculation. A manifest format is the way a digest, filename and sometimes size or chunk evidence are written down. GNU SHA256SUMS, BSD output, CSV and NourForge manifests can all carry SHA-256 while using different layouts.
No text layout makes SHA-256 cryptographically stronger. The practical differences are interoperability, how safely filenames can be represented, whether paths and sizes are explicit, and whether richer evidence such as chunks is available. Checker detects supported layouts and requires a consistent grammar instead of silently guessing ambiguous data.
Advantages over checking one file at a time
- Automatic detection supports NourForge, GNU, BSD, CSV, tab-delimited and other common SHA-256 arrangements.
- Folder verification reports missing, extra, skipped and warning conditions—not only whether one pasted digest matches.
- Expected and actual sizes, hashes, chunk counts, paths and match modes remain available as reviewable evidence.
- Manual file and manual SHA-256 modes support targeted checks without forcing a full manifest workflow.
- Queues, workspaces, detached summaries, detailed results and activity logs help repeat and document larger verification work.
Find a file when only its SHA-256 remains
A filename can change without changing file content. In Manual SHA-256 folder search, Checker hashes the selected scope and finds zero, one or multiple exact content matches for the digest. This is useful when the original filename or location is lost.
Orphan Resolution applies the same principle to unresolved manifest entries after normal path matching. A unique hash match can pair a renamed or moved file; multiple identical candidates are reported for human review rather than treated as certainty about the original path.
Match % answers a different question
Whole-file SHA-256 answers: is the current file identical to the reference? Enhanced Match % answers: after a whole-file failure, what proportion of the recorded 16 MiB reference chunks still match?
A high Match % can indicate a localised difference and a low value broader change, but it is diagnostic—not a measure of playback quality, repairability or cryptographic confidence. It is available only when complete compatible enhanced evidence exists.
COMPARISON
Choosing a verification workflow
| Workflow | Strength | Trade-off |
|---|---|---|
| PowerShell Get-FileHash | Direct, trusted SHA-256 calculation for a known file | Manual comparison and scripting are needed for folders, missing files and reports |
| SHA256SUMS-style command | Portable, established batch list | Layout and path handling vary; results may be terse and usually remain whole-file only |
| General checksum utility | Convenient whole-file checking | Supported formats, ambiguity handling and audit detail vary by tool |
| Archive Verify Checker | Broad format detection, scope-aware results, orphan resolution, queues and optional chunk analysis | Enhanced Match % requires compatible data created in advance |
Standards and technical references
These links provide primary technical context. They do not imply endorsement of NourForge.