Quick answer
Re-save the offending JPEG in any photo editor (Paint, Photos, or GIMP) as a standard JPEG — this strips the unsupported fractional sampling and fixes error 0XC00D103B instantly.
What's actually happening
Error code 0XC00D103B (full name NS_E_WMP_JPG_FRACT_SAMPLE_NOTIMPL) appears when Windows Media Player or any app using the Windows Imaging Component tries to decode a JPEG that uses fractional chroma subsampling. This isn't a corrupt file — it's a JPEG saved with a non-standard encoding that Microsoft's decoder simply never implemented. The JPEG specification allows subsampling ratios like 4:2:2 or 4:2:0, but some cameras or software (notably older Canon DSLRs and some web tools) write files with a fractional ratio — think 4:2:2.5 instead of 4:2:2. Microsoft's decoder chokes on that because it only expects integer values. The image is fine, the decoder isn't. The fix is to make the image standard.
How to fix it (main fix)
- Open the JPEG in Microsoft Paint (Windows 10/11): right-click the file → Open with → Paint.
- Press Ctrl+S or File → Save to overwrite the same file. Paint automatically re-encodes it as a baseline JPEG with standard 4:2:0 subsampling.
- Try opening the file again in Windows Media Player. It works now.
If you don't have Paint (Windows N editions), open it in Photos app, click the three dots → Save as → pick JPEG, then use that new file.
Why this works
Microsoft Paint doesn't preserve exotic JPEG metadata or non-standard subsampling — it strips everything down to a baseline JPEG. The reason step 2 works is that Paint's encoder always writes standard 4:2:0 chroma subsampling (the most compatible form). You lose nothing visually; the image data stays identical. You might lose EXIF data (camera info, GPS), but that's a trade-off for compatibility. If EXIF matters to you, use a more advanced tool like GIMP or IrfanView that can re-encode while keeping EXIF.
Alternative fix if Paint doesn't help
If the error persists after re-saving in Paint, the file might have genuine corruption beyond just subsampling. Try these in order:
- Use ImageMagick (command line): Run
magick convert bad.jpg -sampling-factor 4:2:0 fixed.jpg— this forces integer subsampling and fixes even stubborn files. - GIMP: Open the file, export as JPEG, and under Advanced Options set Subsampling to 4:2:0 (or 4:4:4 for zero subsampling).
- Switch to VLC or MPC-HC: These use their own decoders and aren't affected by this Windows bug. Not a fix but a workaround.
Prevention
This error comes from files created by specific software — old Canon EOS Utility (pre-2010), some PHP image libraries with buggy GD extensions, or web scrapers that saved JPEGs from poorly encoded sources. To avoid it in the future:
- When editing images for Windows, always use "Save for Web" options — they enforce standard encoding.
- If you're a developer serving JPEGs to Windows users, never set custom sampling factors in your image library. Stick with default 2x2 (4:2:0).
- Batch-convert older camera JPEGs using IrfanView's batch mode with "Save as JPEG" and quality 95 — that normalizes them.
Why Microsoft never fixed this
Microsoft's decoder was written in the early 2000s and never updated to handle fractional subsampling. The JPEG committee's spec technically allows it, but no mainstream encoder uses it. Microsoft decided the engineering cost wasn't worth it for zero practical benefit. The real question is why your file has it — that's the software that created it being non-standard, not Windows being broken. Point the finger at the encoder, not the decoder.