Media information and access checks
iCouldPlay media inventory: separate container, video codec and audio streams
Build a layered inventory for a recoverable sample and explain playback differences from stream information rather than an MP4 suffix.
On this page
Two MP4 files can contain different video and audio codecs, and visible pictures do not validate every audio stream. Play a small owned sample in iCouldPlay and record container, video, audio and subtitles separately. This is not a format ranking or a guarantee that every listed codec combination works on every device.
Identify one recoverable file
Choose a self-recorded or permitted short clip, retain its original outside the app and record name, size, duration and source. Import normally and verify identity, then note app/system versions and any visible engine state. Do not start by comparing a processed remote copy against a local original, changing network and encoding together. Inventory one file without transcoding a library or using unfamiliar test media.
Separate the suffix from the container
Record the suffix separately from a trusted tool’s container report. Renaming MOV to MP4 neither remuxes the file nor changes video coding. Use visible media-information fields if offered; otherwise use an existing trusted local tool read-only, naming it as external. Leave unavailable fields unknown instead of inferring a container from successful opening or installing an unfamiliar codec pack to complete the sheet.
Record video type and dimensions separately
For each video stream, record codec, dimensions and other fields actually supplied, distinguishing the main moving image from an attached cover image. H.264 and HEVC differ, yet one codec name does not explain every compatibility result. Dimensions such as 1920×1080 are pixel extents, not bitrate, file size or frame rate. Preserve the average-rate field’s identity without turning it into a per-image timing map.
Identify audio streams individually
List each audio stream’s index, codec, language and channel information, marking missing tags rather than treating stream one as primary dialogue. A sample may have only one audio track or may include commentary; record actual streams without inventing extras to satisfy the exercise. Verify language by listening later. Channel metadata does not establish how many independent directions headphones output, and successful playback is not certification of a home-theatre layout or audio effect.
Keep subtitles and cover art distinct
List embedded subtitle streams, external subtitle files and cover attachments separately with their sources. Embedded streams, separate files and burned-in text are different objects despite all producing visible text. A cover attachment is not a second film. Distinguish not found from not exposed by a tool, then compare visible app selection lists rather than reporting total stream count as video count.
Check playback at three content anchors
Play the opening, then start before middle and near-end anchors and record picture, sound and selected captions separately. A good opening does not prove later integrity. Missing captions need not mean video decode failure. Tie observations to a stream or file rather than one “MP4 works” verdict. Start earlier when seeking is coarse, without assuming frame-step controls, and retain initial failures alongside later success.
Change one condition rather than rename
If another engine is visibly available, compare it once on the same file and anchor, preserving the original choice. Without the control, retain current conditions. For server media, establish whether a processed copy arrived; do not assign a processed copy’s metadata to the local original. Do not disguise conversion through renaming or overwrite unbacked originals. Finish the original’s inventory before any external conversion, give the new output a separate identity and information record, and verify retained audio/subtitles, not merely opening success.
Keep metadata limits visible
A metadata tool reports recognised structure without proving every block decodes or the visual content is correct. Record tool, identity and a shareable error summary without private full paths. Official ffprobe documentation distinguishes container and stream information; installation or commands are not required here. Do not invent unknown fields or verify HDR/surround effects solely from a label.
Save the layered result and revisit
Retain suffix, container, stream identities, three playback results and unknowns. Reopen the same file and revisit one anchor. State observed picture/sound and untested tracks without generalising to every file with that suffix. Restore normal selections and keep original, report and notes separately. This is an inventory and playback baseline, not exhaustive device acceptance, performance testing or a safety guarantee for arbitrary media.
Related tutorials and sources
Engine checks, Original and processed copies.
Checked on November 8, 2026 against the official iCouldPlay App Store listing and Apple public lookup: version 1.1.1, requiring iOS / iPadOS 15.0 or later. The public 1.1.1 release notes list Kodi, OpenList / AList and S3-compatible storage, with improvements to source switching and captions after seeking. The tasks use this public scope and existing playback capabilities. Check controls in your installed version. Self-made samples, manual records and external file preparation are workflows, not newly promised automatic features. No physical-device or private-server test is claimed.