Format reference
TIFF
TIFF stores raster image data and descriptive fields in a tag-based, extensible file structure.
Quick reference
- Family
- Tag-based raster container
- Extensions
- .tif · .tiff
- MIME type
- image/tiff
- Identification
- 49 49 2A 00 — little-endian classic TIFF · 4D 4D 00 2A — big-endian classic TIFF; BigTIFF has different headers
- Compression
- Uncompressed or compression schemes such as PackBits, LZW and CCITT; extensions add other variants
- Alpha / transparency
- Extra samples can describe alpha; not universal across TIFF readers
- Animation / multiple images
- Multiple images/pages, not inherently a timed animation
TIFF stores raster image data and descriptive fields in a tag-based, extensible file structure.
Directories, tags and several kinds of raster data
Classic TIFF declares byte order and points to image-file directories. Tags describe dimensions, samples, compression and data locations. Multiple directories can represent several images or pages; TIFF is not a page-description language like PDF.
Compression varies by file and application. A TIFF may be uncompressed or use a supported compression scheme, so “TIFF” alone does not mean lossless or compatible. Extra samples can carry alpha; subdirectories and private tags can add information beyond a simple raster.
Metadata can extend beyond familiar tags
TIFF workflows can contain EXIF, GPS, IPTC, XMP, ICC, MakerNotes, thumbnails and subIFDs. Extensions and private tags make inspection coverage important: reading familiar metadata does not establish that every private structure was decoded.
The narrower PRVKIT adapters
Image Inspector accepts TIFF and reports supported Metadata Engine V2 groups. Supported thumbnails are reported where the parser supplies them; arbitrary subIFD/private-field completeness is not claimed.
The Editor’s TIFF adapter is narrower: a Pro-gated import for supported uncompressed 8-bit grayscale/RGB/RGBA, with selected pages imported as raster layers. It is not general TIFF compression support or TIFF export. Convert Image and Remove Metadata do not accept TIFF directly; convert in a compatible workflow before using those raster tools.
Preserve the source when changing containers
When one page becomes PNG or JPEG, other pages and directory relationships are not represented in that output. JPEG adds a lossy step; PNG stores the decoded pixel grid. Metadata and source color information require separate preservation decisions.
Do not assume that a TIFF-compatible metadata inspector is a TIFF editor or encoder. A specialized workflow may be needed for unusual sample types, compression, private tags or long-term preservation requirements.
Agree on a TIFF profile, not just an extension
TIFF is flexible enough that two files with the same ending can place very different demands on a reader. Byte order, sample arrangement, compression and directory relationships all matter. An application supporting a baseline uncompressed page should not claim to support every scanner, camera or specialist TIFF variant.
For a professional handoff, ask what the receiving workflow expects: one page or several, a particular compression, a sample depth and any required color information. A successful thumbnail display is not proof that every page or channel has been decoded. Keep the original while testing the actual recipient's application.
The common classic headers identify byte order and the TIFF family, but format recognition is only the beginning. BigTIFF uses a related but different structure for larger offsets. A familiar signature does not remove the need to validate directory counts, data ranges and resource requirements before processing.
Multipage is not the same as PDF-style page editing
Several image directories can represent pages or other related images. Their presence does not automatically create a timed animation or the page-description model of a PDF. A converter that exports one image may omit other directories unless it explicitly supports selecting and preserving them.
A scan set can therefore lose meaning if one rendered preview is presented as the whole document. Check the expected number of pages and what the actual tool imports. PRVKIT's narrower Editor TIFF adapter and supported single-page Converter path should not be described as a universal multipage conversion service.
When the recipient needs one visual page, a deliberate PNG or JPEG copy may be appropriate. When they need every page and its relationships, use a workflow qualified for that purpose. Keep the source so that a later requirement does not have to be reconstructed from an incomplete delivery image.
Compression and sample depth are independent questions
A TIFF can use uncompressed data or one of several compression schemes. The name alone does not establish that it is lossless, small or easy to decode. A reader may support one scheme and reject another even when width and height appear ordinary.
Sample depth and color representation also matter. A high-bit-depth photographic master is not equivalent to an eight-bit preview. A later PNG or JPEG export through an ordinary browser surface cannot be assumed to preserve every original sample or profile merely because its visible thumbnail looks familiar.
PRVKIT's current TIFF import boundary is deliberately narrower than the format's full possibilities. Respect that boundary rather than changing the extension or substituting a preview for a claimed full-fidelity import. For archival or color-critical work, retain the original and verify through an appropriate receiving application.
Metadata can be useful and incomplete at once
Tags can describe ordinary image properties alongside EXIF, GPS, descriptive fields, profiles, thumbnails and private extensions. An inspector can report many useful fields without understanding every subdirectory or proprietary value. “Not detected” is a statement about supported coverage, not proof that the file has no other information.
A supported thumbnail can help identify a file, but it may differ from a main image or represent only part of a larger document. Do not treat the preview as complete evidence of what every page contains. Visible identifiers also remain separate from the metadata report.
Choose a derived copy for its actual role
A PNG can be useful for a decoded image needing transparency or crisp edges. A JPEG can suit an opaque photographic delivery, with a new lossy step. A PDF may be appropriate for a document handoff prepared through a qualified page workflow. None of those labels automatically preserves the TIFF's directory relationships or original high-depth data.
Open the actual saved output and check dimensions, page selection, orientation and important detail. If metadata or color handling matters, verify those separately. A started download is not proof of complete saving. Use Image Inspector for supported TIFF fields and File Inspector & Hash for byte comparison with a trusted reference, keeping each result within its real scope.
Metadata inspection coverage
prvkit tools
Related reading
Technical sources
- IANA image/tiff — Registered media type
- TIFF 6.0 specification — Endian headers, directories, compression and extra samples
- TIFF media type registration — TIFF variants and media type