File format glossary

Information describing pixel transparency or opacity in an image. Formats such as PNG and WebP can store an alpha channel, allowing pixels to be fully or partially transparent. JPG does not support an alpha channel - transparent pixels must be placed over a solid background when converting to JPG.

The relationship between an image's width and height, expressed as two numbers separated by a colon. Common examples include 16:9 (widescreen), 9:16 (portrait/vertical), 1:1 (square) and 4:5 (portrait social). Cropping or padding is required when fitting an image into a different aspect ratio.

AV1 Image File Format. A modern image format designed for efficient compression. AVIF can produce smaller files than WebP or JPG at comparable visual quality. Browser and application compatibility can vary, so checking support before deploying AVIF widely is advisable.

An image made from a fixed grid of pixels, where each pixel stores color information. Common raster formats include JPG, PNG, WebP, GIF and BMP. Raster images lose quality when enlarged beyond their original pixel dimensions because no additional detail exists to fill the extra space.

Bitmap image format. A traditional raster format that stores pixel data with minimal compression, which can result in relatively large file sizes. BMP is rarely used for web delivery but remains supported by many desktop applications.

A browser drawing surface provided by the HTML Canvas API. Many client-side image operations - such as resizing, format conversion and compositing - use the canvas to read pixel data and produce a new image result without sending the file to a server.

Reducing the amount of data required to store or transmit a file. Lossless compression preserves all encoded information so the original can be reconstructed exactly. Lossy compression intentionally removes some information to achieve smaller sizes, which cannot be fully reversed.

Visible distortions introduced by aggressive lossy compression. Common examples include blockiness around edges, ringing halos near sharp transitions, smearing of fine details and loss of texture. Artifacts become more noticeable at lower quality settings and can accumulate with repeated lossy re-encoding.

The width and height of an image measured in pixels. For example, 1920 × 1080 means 1920 pixels wide and 1080 pixels tall. Dimensions describe the pixel grid size and are different from file size - two images with identical dimensions can have very different file sizes depending on format and compression.

Dots per inch. A measurement primarily meaningful in print workflows, describing how many ink dots a printer places per inch of paper. Changing a DPI metadata value stored in an image file does not add pixel detail to the image - it only changes the number used to calculate print dimensions.

Pixels per inch. Describes how many pixels are displayed per inch on a screen or printed per inch on paper. PPI is related to but distinct from raw pixel dimensions. A 300 PPI print from a 3000 × 3000 image produces a 10-inch square print; the same image at 72 PPI would produce a much larger but lower-density print.

Exchangeable Image File Format. Metadata commonly stored inside photographs by cameras and phones. EXIF data may include camera make and model, capture date and time, lens settings, orientation flag and sometimes GPS location coordinates. Removing EXIF data can reduce this embedded information before sharing.

Graphics Interchange Format. A raster image format commonly used for simple animation. GIF supports multiple frames played in sequence, a limited 256-color palette per frame and basic transparency. Animated GIF files can become very large because each frame stores pixel data separately.

High Efficiency Image Container / High Efficiency Image Format. An efficient image container commonly used by modern iPhones and Android devices. HEIC can store high-quality photographs at roughly half the file size of equivalent JPG images. Some applications and upload forms still require JPG or PNG.

Joint Photographic Experts Group. JPEG and JPG refer to the same image format - the difference is only the file extension length used on older systems. JPG is best suited to photographs and uses lossy compression, meaning some image data is discarded to reduce file size. It does not support transparency.

Compression that preserves all encoded information so the original data can be reconstructed exactly from the compressed form. PNG uses lossless compression for its pixel data. Lossless does not mean small - a losslessly compressed image can still be a large file depending on its content.

Compression that intentionally removes some information to achieve smaller file sizes. The original data cannot be perfectly reconstructed. Repeated lossy re-encoding of the same image can introduce additional degradation each time. JPG and lossy WebP use lossy compression.

Information stored about a file separately from its primary visible or audible content. Image metadata may include camera details, timestamps, software used and location. PDF metadata may include title, author and creation date. Metadata can be read, edited or removed without changing the main content.

The smallest addressable sample in a raster image. Each pixel stores color information, typically as red, green and blue values. The total number of pixels in an image is its width multiplied by its height.

Portable Network Graphics. A lossless raster format commonly used for transparent graphics, screenshots, logos and line art. PNG supports an alpha channel for transparency. Because it is lossless, PNG files of photographs are typically larger than equivalent JPG files.

An encoder setting that influences the trade-off between compression and visible quality. A higher quality value generally produces a larger file with fewer artifacts. An 80% quality setting in one encoder is not guaranteed to produce the same visual result or file size as 80% in a different encoder.

For digital images and video, resolution normally refers to pixel dimensions - for example, 1920 × 1080. In print contexts, resolution may also refer to DPI or PPI. Higher pixel dimensions allow more detail to be stored but also produce larger files.

Scalable Vector Graphics. A vector image format based on mathematical shapes, paths and text rather than a fixed pixel grid. SVG images can be scaled to any size without the pixelation that affects raster images. SVG is well suited to logos, icons and diagrams.

Pixels that are partly or fully invisible, allowing the background behind the image to show through. Transparency is stored using an alpha channel. PNG and WebP support full alpha transparency. GIF supports binary (on/off) transparency. JPG does not support transparency.

Artwork represented using mathematical paths, shapes and curves rather than a fixed pixel grid. Vector images can be scaled to any size without the pixelation that affects raster images. SVG is the primary vector format used on the web.

A modern raster image format developed by Google. WebP supports lossy compression, lossless compression, transparency and animation. It typically produces smaller files than equivalent JPG or PNG images. Browser support is now broad across current versions of Chrome, Edge, Firefox and Safari.

Portable Document Format. A file format designed to present documents consistently across different devices and operating systems. A PDF may contain text, vector graphics, raster images, scanned pages, form fields and metadata. The visual layout is preserved regardless of the viewer's fonts or settings.

A standard paper size measuring 210 × 297 mm (approximately 8.27 × 11.69 inches). A4 is the most widely used document page size internationally and is the default in most countries outside North America.

A common North American document page size measuring 8.5 × 11 inches (approximately 216 × 279 mm). US Letter is the default page size in the United States and Canada.

Combining two or more PDF documents into a single PDF file. The resulting document contains all pages from the source files in the specified order. The original files are not modified.

Separating a multi-page PDF into smaller PDF documents or individual pages. Splitting is useful for distributing specific sections of a document or reducing file size.

Creating a new PDF containing only selected pages from a source document. The original document is not modified. Extracting is useful when only a portion of a larger document is needed.

Creating a new PDF that omits specified pages from the source document. The original document is not modified. The result contains all pages except those selected for removal.

Changing the sequence of pages within a PDF document. A new PDF is created with pages arranged in the specified order. The original document is not modified.

Changing the orientation of one or more pages within a PDF - for example, from portrait to landscape. Rotation changes the page orientation stored in the document structure without intentionally recompressing the visible page content.

Descriptive document properties stored within a PDF file, such as title, author, creator application, producer and creation or modification dates. PDF metadata can be read, edited or removed. Metadata fields are not proof of authorship because they can be changed by anyone with access to the file.

A unit of digital storage equal to 1,000 bytes in decimal measurement (SI). Kilobytes are commonly used to describe small file sizes. Note that some systems use kibibytes (KiB, 1,024 bytes) instead, which can cause slight differences in displayed sizes.

A unit of digital storage equal to 1,000,000 bytes (1,000 KB) in decimal measurement. Megabytes are commonly used to describe medium-sized files such as photographs, audio clips and documents.

A unit of digital storage equal to 1,024 bytes using binary measurement. KiB is the technically precise binary equivalent of KB. Some operating systems and tools report file sizes in KiB while labelling them as KB, which can cause apparent discrepancies.

A unit of digital storage equal to 1,048,576 bytes (1,024 KiB) using binary measurement. MiB is the technically precise binary equivalent of MB. The difference between MB and MiB becomes more noticeable at larger file sizes.

A container file used to package one or more files, often with compression applied. Archives are useful for distributing multiple files as a single download, reducing transfer size and preserving folder structure.

A widely supported archive format capable of storing multiple files and folders in a single container. Compression may be applied individually to each entry. ZIP is supported natively by most operating systems without additional software.

A compression format that applies the DEFLATE algorithm to a single data stream. GZIP normally produces a single compressed file with a .gz extension. Unlike ZIP, GZIP is not designed to bundle multiple files - it compresses one stream at a time.

The relationship between the original uncompressed size and the compressed size of a file or archive. A higher compression ratio means the compressed result is smaller relative to the original. Already-compressed files such as JPG, MP3 and MP4 typically show little additional size reduction when placed in a ZIP.

Restoring the files or data contained inside an archive to their original form. Extracting a ZIP or GZIP file produces the original files that were stored inside.

A lossless data compression method commonly used inside ZIP and GZIP files. DEFLATE combines LZ77 compression with Huffman coding. It is the standard compression algorithm for most ZIP archives.

A calculated value derived from a file's data, used to compare files and detect changes or corruption. If a file is modified - even slightly - its checksum will differ from the original. Checksums are used to verify that a downloaded file matches the publisher's trusted copy.

A fixed-length digest generated from input data using a cryptographic hash function. Cryptographic hashes are designed to be one-way (the original data cannot be recovered from the hash) and collision-resistant (it should be computationally infeasible to find two different inputs that produce the same hash).

Another term commonly used for the output value of a hash function. The digest is a fixed-length string that represents the input data. "Hash" and "digest" are often used interchangeably.

Secure Hash Algorithm 256-bit. A modern cryptographic hash algorithm that produces a 64-character hexadecimal digest. SHA-256 is widely used for file integrity verification and is the recommended choice when a trustworthy modern checksum is available.

Message Digest 5. An older hash algorithm that produces a 32-character hexadecimal digest. MD5 should not be relied upon for modern collision-resistant security because known weaknesses allow deliberate collisions to be constructed. It remains useful when a legacy workflow specifically publishes an MD5 checksum.

Cyclic Redundancy Check 32-bit. A short checksum useful mainly for detecting accidental data corruption, commonly used inside ZIP archives. CRC-32 is not cryptographically secure and should not be used as a security integrity mechanism.

Hash-based Message Authentication Code. A construction that combines a secret key with a cryptographic hash function. HMAC allows a party who possesses the same secret key to verify both the integrity and authenticity of a message. Unlike a plain hash, HMAC requires knowledge of the shared secret to verify.

A situation where two different inputs produce the same hash value. Collision resistance is a critical property for cryptographic hash functions - if collisions can be deliberately constructed, an attacker could substitute a malicious file that produces the same hash as a trusted file.

Confidence that a file has not changed relative to a trusted reference. File integrity is typically verified by comparing a checksum or cryptographic hash of the received file against a hash published by the trusted source. A matching hash confirms the bytes are identical but does not prove the publisher is trustworthy.

The amount of encoded data used per second of media, measured in kilobits per second (kbps) or megabits per second (Mbps). Higher bitrate generally allows more detail to be preserved but increases file size. Bitrate applies to both video streams and audio streams.

A method used to encode and decode media data. The codec determines how video or audio is compressed and stored. Common video codecs include H.264, VP8 and VP9. The codec is separate from the container format - for example, an MP4 container can hold video encoded with different codecs.

A file format that holds one or more media streams (video, audio, subtitles) together with metadata. Examples include MP4 and WebM. The container and the codec are not the same thing - an MP4 container does not specify a single codec.

The number of individual frames displayed each second in a video. Common frame rates include 24 fps (cinema), 30 fps and 60 fps. Higher frame rates produce smoother motion but require more data.

An independently encoded video frame that does not depend on surrounding frames for its data. Keyframes serve as reference points for the frames between them. Seeking to a non-keyframe position in a video requires decoding from the nearest preceding keyframe.

A popular media container format (MPEG-4 Part 14) that can hold video, audio and subtitle streams. MP4 is widely supported across devices and platforms. MP4 is a container, not a specific codec - the video quality and compatibility also depend on the codec used inside the container.

An open media container format designed for web and browser workflows. WebM typically uses VP8 or VP9 video codecs and Vorbis or Opus audio. Browser-based video encoding tools often produce WebM because browsers expose compatible encoding APIs for this format.

A common video resolution label associated with approximately 1280 × 720 pixels. The "p" stands for progressive scan. 720p is considered HD (high definition) and is widely supported across devices and streaming platforms.

A common video resolution label associated with approximately 1920 × 1080 pixels. Also called Full HD. 1080p is a standard resolution for streaming, broadcast and consumer video production.

A common consumer ultra-high-definition video resolution around 3840 × 2160 pixels (four times the pixel count of 1080p). Also referred to as UHD. 4K files are substantially larger than 1080p files at equivalent quality settings.

A still image captured or created to represent a video. Thumbnails are used as preview images on video platforms, in search results and in media players. A thumbnail is separate from the video file itself.

MPEG-1 Audio Layer III. A widely supported lossy compressed audio format. MP3 reduces file size by discarding audio information that is less perceptible to human hearing. It is the most common format for music distribution and general audio sharing.

Waveform Audio File Format. An audio container commonly used to store PCM (uncompressed) audio. WAV files are typically much larger than MP3 files but contain no lossy compression, making them useful for editing and production workflows where maximum source quality is needed.

Pulse-Code Modulation. A common uncompressed digital representation of audio, where the amplitude of the audio signal is sampled at regular intervals and stored as numerical values. PCM audio is the basis of WAV files and audio CDs.

The number of audio samples captured or represented per second, measured in hertz (Hz) or kilohertz (kHz). Common sample rates include 44.1 kHz (CD audio) and 48 kHz (professional video/broadcast audio). Higher sample rates can represent higher audio frequencies.

The number of bits used to represent each audio sample in PCM audio. Common bit depths are 16-bit (CD quality) and 24-bit (professional recording). Higher bit depth provides greater dynamic range.

The amount by which an audio signal level is increased or decreased. Applying positive gain makes audio louder; negative gain makes it quieter. Excessive gain can push the signal beyond the available range, causing clipping.

The highest instantaneous level reached by an audio waveform. Monitoring peak levels helps prevent clipping. A peak level at or above 0 dBFS (decibels relative to full scale) in digital audio indicates clipping.

Distortion that occurs when an audio signal level exceeds the available range. In digital audio, clipping produces a harsh, distorted sound because the waveform is cut off at the maximum value. Clipping cannot be repaired after it has occurred in the source recording.

Adjusting audio level relative to a target peak or loudness value. Peak normalization raises or lowers the entire audio so the loudest point reaches a specified level. Normalization adjusts overall level but does not repair noisy, distorted or damaged recordings.

JavaScript Object Notation. A structured text format for representing data using objects (key/value pairs), arrays, strings, numbers, Booleans and null. JSON is widely used for data exchange between applications and APIs.

A collection of key/value pairs enclosed in curly braces in JSON. Keys must be strings. Values can be strings, numbers, Booleans, null, arrays or nested objects. Example: {"name": "example", "count": 3}

An ordered collection of values enclosed in square brackets in JSON. Values can be of any JSON type. Example: ["apple", "banana", 42, true]

Checking whether a piece of text follows valid JSON syntax - correct use of braces, brackets, quotes, commas and colons. Syntax validation confirms the text is parseable JSON but does not check whether the data follows a specific business schema or JSON Schema definition.

Formatting structured text such as JSON or XML using consistent whitespace and indentation to make it easier for humans to read. Pretty-printed JSON is functionally identical to compact JSON - the formatting does not change the data.

Comma-Separated Values. A tabular text format where each line represents a row and fields within each row are separated by a delimiter character, typically a comma. CSV is widely used for spreadsheet data, database exports and data interchange.

A character used to separate values in a structured text format. In CSV, the comma is the standard delimiter, though tabs, semicolons and other characters are also used. Fields containing the delimiter character must be quoted or escaped.

A method of representing characters that otherwise have special meaning in a format. In JSON, a double-quote inside a string value must be escaped as ". In CSV, a field containing a comma must be enclosed in double quotes.

Extensible Markup Language. A structured markup format using elements (opening and closing tags) and attributes to represent hierarchical data. XML is used in many data exchange formats, configuration files and document standards.

A component of an XML document consisting of an opening tag, optional content or child elements, and a closing tag. Example: <title>My Document</title>. Elements can be nested to represent hierarchical data.

Name/value information contained within an XML opening tag. Example: <image width="800" height="600">. Attributes provide additional properties for an element without creating child elements.

A reversible encoding scheme that represents binary data as printable ASCII text using 64 characters. Base64 is used to embed binary data (such as images) in text-based formats. Base64 is not encryption - anyone who receives a Base64-encoded value can decode it without a key.

A widely used variable-width Unicode text encoding. UTF-8 can represent every character in the Unicode standard while remaining backward-compatible with ASCII for the first 128 characters. UTF-8 is the dominant encoding for web pages and text files.

Processing performed by code running in the visitor's browser rather than on a remote server. Client-side processing means the computation happens on the visitor's device using the browser's JavaScript engine and available APIs.

As used by NEXDOWNLOAD: the selected source file is processed on the visitor's device for supported workflows rather than being sent to a NEXDOWNLOAD conversion server. Processing speed depends on the device's available memory and CPU. Local processing does not make a browser-based website equivalent to offline software - ordinary web requests, analytics and advertising may still occur.

A browser object representing binary data. In browser-based file tools, a Blob may hold the processed result - such as a converted image or compressed archive - before the user downloads it. Blobs exist in browser memory and are not automatically saved to disk.

A temporary browser URL (beginning with blob:) that references locally held data such as a Blob. Object URLs allow the browser to treat in-memory data as a downloadable resource. They are temporary and become invalid when the page is closed or the URL is explicitly revoked.

The temporary memory available to a browser tab for running JavaScript and holding data. Large files require more browser memory to decode, process and encode. Available memory varies by device and browser. Insufficient memory can cause processing to fail or the browser tab to crash.

Whether the browser and operating system can decode or encode a particular media codec. Browser codec support varies between browsers, operating systems and device types. A file that plays correctly in one browser may not be decodable in another if the required codec is not available.