X3F to TS Converter

How to Convert X3F to TS Online

1. Upload Your X3F File: Drag and drop or click "+ Add Files" to load Sigma Foveon RAW files from SD9/SD10/SD14/SD15, the DP and dp Quattro compacts, or the sd Quattro/sd Quattro H. Batch upload works — every X3F will become one frame (or one short clip) of the output TS.
2. Pick Merge Strategy and Quality Preset: Choose "Merge images" to stitch all uploaded X3F frames into one continuous transport stream, or "Video per image" to emit a separate .ts per file. Set Quality Preset to Very High (recommended for archival) or Constant/Constraint Quality for a fixed CRF. Output is MPEG-2 video + MP2 audio by default — the safe baseline for DVB and legacy hardware decoders.
3. Set Duration, Resolution and Background (Optional): Pick a per-frame Duration (1/60-second up to 10 seconds — 5 s is the default), keep original Foveon resolution (5424×3616 from a Quattro, 4704×3136 from an SD1 Merrill) or down-sample to a Preset Resolution (4K, 1440p, 1080p, 720p) for HLS-friendly segments. Set Background Color (default black) for letterbox padding when aspect ratios differ.
4. Convert and Download: Click "Convert". Files process server-side in your browsing session — no Sigma Photo Pro install, no watermark, no sign-up. The resulting .ts plays in VLC, MPC-HC and FFmpeg-based players, or drops directly into an HLS .m3u8 segment list.

Why Convert X3F to TS?

X3F is Sigma's proprietary container for Foveon X3 sensor RAW data — three vertically stacked photodiode layers per pixel, no Bayer demosaic. The format is unmatched for color fidelity but unsupported almost everywhere outside Sigma Photo Pro. MPEG-TS (.ts) is the opposite: a 1995-vintage ISO/IEC 13818-1 container designed for lossy transmission over satellite, terrestrial broadcast and IPTV, with 188-byte packets that resync after dropped bits. Converting an X3F sequence to TS turns a stack of orphaned RAWs into a streamable video — useful for:

• HLS slideshow segments — Apple HLS uses .ts segments by default (RFC 8216), typically 6-second chunks per Apple's authoring spec. Convert a folder of Foveon stills to a 6-second TS and drop it straight into a .m3u8 playlist.
• DVB/ATSC broadcast graphics — station IDs, weather slides and lower-thirds for over-the-air channels expect MPEG-2 in TS; X3F→TS produces a feed-ready clip without re-muxing.
• Foveon time-lapse archival — sd Quattro shooters using the camera for filmic 8K timelapse get a transport-stream master that survives partial corruption better than an MP4 — TS packets resync; MP4 atoms do not.
• Concatenation without re-encode — TS files can be joined with cat a.ts b.ts > out.ts, so per-shot exports become a single archival reel via a one-line shell command.
• IPTV / set-top-box playback — many older networked players (D-Link DSM, WD TV Live, Linux satellite STBs) reject MP4/MOV but accept MPEG-2 TS over UDP or HTTP.
• Legacy AVCHD camcorder workflow — AVCHD's .m2ts is a TS variant with 4-byte timecode prefixes; if you're cutting a Foveon B-roll slideshow into an AVCHD project, MPEG-TS is the closest hand-off.

X3F vs TS — Format Comparison

Codec & Quality Quick Guide

Frequently Asked Questions

Why convert a Sigma Foveon RAW to a video transport stream at all?

X3F holds full three-layer sensor data — gorgeous, but unplayable in any browser, NLE timeline preview window, or HLS player. Wrapping a sequence of X3F frames into MPEG-TS gives you something a CDN, set-top box, or <video> tag can actually stream. Common workflows: Foveon timelapses destined for HLS, Sigma sd Quattro slideshows for DVB station playout, and concatenation-friendly archival reels (TS files merge with plain cat, MP4 doesn't).

Will the converter debayer my Foveon X3 data correctly?

There is no "debayer" — Foveon stacks red, green and blue photodiodes vertically, so each pixel already has full RGB without interpolation. The conversion reads the FOVb-signed X3F payload and renders it to standard 8-bit RGB before encoding to MPEG-2 / H.264 / H.265. Color rendition won't match what Sigma Photo Pro produces from the same RAW (SPP applies Sigma's tuned color matrix), but it's correct, neutral, and consistent across a batch.

What's the difference between "Merge images" and "Video per image"?

"Merge images" concatenates every uploaded X3F into a single .ts at the duration-per-frame you pick — useful for HLS segments, slideshows and timelapses. "Video per image" emits one .ts file per X3F, each holding a single still held for the chosen duration — useful when you need separate segments for adaptive bitrate streaming or per-frame DVB graphics.

Why MPEG-TS instead of MP4 for an image slideshow?

Three reasons. (1) TS is HLS's native segment format per RFC 8216 — drop the .ts into an .m3u8 playlist and you're done. (2) TS resyncs at the next 188-byte sync byte after a transmission error; MP4's moov atom is a single point of failure. (3) Broadcast pipelines (DVB, ATSC, AVCHD camcorders' .m2ts) speak MPEG-TS natively. If you don't need any of these, convert X3F to MP4 instead.

Which Sigma cameras shoot X3F that this tool will accept?

Any X3F since the 2002 SD9 — SD10, SD14, SD15, DP1/DP2/DP3 Merrill, dp1/dp2/dp3 Quattro, the SD1 and SD1 Merrill DSLRs, and the sd Quattro and sd Quattro H mirrorless bodies. Resolutions range from ~3.4 MP-per-layer SD9 files to the 5424×3616 (39 MP Bayer-equivalent) Quattro RAWs. The newer Sigma fp / fp L cameras shoot DNG, not X3F — for those, use DNG to TS.

What output resolution should I pick for HLS streaming?

For HLS: 1920×1080 at H.264 CRF 23 is the safe default — every iOS device, smart TV, and HLS-capable browser handles it. For ladders, output once at 1080p and once at 720p so the player can switch on bandwidth changes. Don't stream the native 5424×3616 Foveon resolution — almost no playback device benefits, and the bitrate balloons. For archival masters where decoded quality matters more than streaming, keep "Original" resolution and pick the Very High preset.

Will the audio track be empty?

Yes by default — X3F is a stills format with no audio payload, so the output TS contains only video (and silent MP2 padding if a multiplexer requires an audio PID). If your downstream pipeline rejects audio-less streams, mux a silent track in FFmpeg post-export: ffmpeg -i in.ts -f lavfi -i anullsrc -c:v copy -c:a mp2 -shortest out.ts.

Can I use this for AVCHD camcorder workflows?

Sort of. AVCHD uses .m2ts, which is MPEG-TS with a 4-byte timecode prefix on each packet (192 bytes instead of 188). The .ts this tool produces is plain MPEG-TS — most AVCHD NLEs (PowerDirector, Vegas, Premiere) ingest it cleanly, but if your editor strictly demands .m2ts, rename and remux with ffmpeg -i in.ts -c copy out.m2ts after download.

How does this compare to converting through Sigma Photo Pro + a separate encoder?

Sigma Photo Pro will give you better color fidelity from the Foveon RAW (it's tuned to the sensor) but only exports stills — you'd then load the TIFFs into an NLE or FFmpeg and encode to TS yourself, often a multi-hour pipeline for a long timelapse. This tool collapses both steps into one upload-and-click flow; the trade-off is generic debayer color vs Sigma's tuned pipeline. For one-off slideshows and quick HLS segments, this is faster; for fine-art color work, develop in SPP first and feed the TIFFs to Image to Video.