MP4 MP4 Analyzer by MediaOnDeck

// faq

Frequently asked questions

Is my video uploaded anywhere?

No. Parsing, decoding and analysis all happen in your browser. There is no upload endpoint. Unplug your network and open a file — it still works, which is the most direct way to verify that processing is truly local.

Do I need an account? Is it paid?

No account required, and all features are currently free.

Which formats and codecs are supported?

Containers: MP4 / MOV / M4V / 3GP / MKV. Video codecs: H.264 (AVC) and H.265 (HEVC). Audio tracks can be inspected at the container and sample-table level, but not at the bitstream level.

Is H.265 / HEVC supported?

Yes, including block-level analysis (CU partitioning, QP, motion vectors). Because HEVC uses quadtree CTU partitioning, block sizes vary far more within a frame than in H.264 — you can see that directly in the view.

What about AV1 or VP9?

Not supported yet. Block-level analysis currently covers H.264 and H.265 only.

Why do some frames show no QP or MV?

A few different reasons, none of them a bug:

  • I frames have no motion vectors. They are intra-coded and reference nothing, so there is no MV — QP is still present;
  • Blocks are too small to draw arrows. Arrows appear only once a block is large enough on screen at the current zoom — zoom in to see them;
  • Audio or other tracks. Audio frames and metadata have no notion of QP or MV.

Why do motion vectors point opposite to the apparent motion?

An arrow shows where this block's content came from in the reference frame, not where it goes. If the picture moves right, the arrows point left.

Is there batch analysis or an API?

Currently it is single-file, interactive analysis — no batch jobs and no public API. If you need batch QC, automated testing or integration with your own system, email us with the use case; that kind of request gets prioritized.

Can I export the results?

Not yet. If you need QP / MV data exported for downstream analysis, let us know by email.

Which browsers are required?

A recent Chrome or Edge is recommended: frame preview uses the browser's built-in hardware decoder. On other browsers, bitstream analysis and block data still work — just without the picture preview.

Is there a file size limit?

No hard limit, but analysis runs in browser memory: larger files and higher resolutions take longer to parse on first load and use more memory.

Why is it slower than desktop tools?

Desktop tools run native code with platform hardware acceleration; the browser version cannot use quite the same shortcuts. You trade some speed for zero install, cross-platform reach, and files that never leave your machine.

Can I use it commercially or embed it in my own system?

Analysis is built on FFmpeg (LGPLv3); as the license requires, the FFmpeg sources used by this project and the full license are published. For enterprise integration, private deployment or custom work, get in touch by email.

How does it compare to Elecard StreamEye, CodecVisa or mp4info?

They do not solve exactly the same problem. Elecard StreamEye Studio and CodecVisa cover more codecs (AV1 / VP9 / VVC) and add buffer (HRD / DPB) analysis, standard compliance verification, and PSNR / SSIM / VMAF against a reference YUV — none of which is possible in a browser, and this tool does not attempt any of it. mp4info is a command-line tool built for scripting and batch work, but it has to be installed first. This tool's niche is block-level visualization (QP heatmap and motion vectors) at zero install cost, which suits quick localization, reviews and teaching. The full comparison is here.

How do I report a problem?

Email liuhang887654@gmail.com with your browser version, the file's key parameters (container / codec / resolution) and what you observed.