ORF to TS Converter

ORF to TS: Putting an Olympus Photograph Into a Transport Stream

ORF is Olympus raw — the undeveloped sensor readout from an OM-D, PEN or OM System body, stored in a TIFF-like container whose header carries Olympus's own IIRO signature instead of TIFF's magic number. TS is the MPEG-2 transport stream defined in ISO/IEC 13818-1: a flat run of fixed 188-byte packets designed so a corrupted packet costs you a fraction of a second rather than the rest of the file.

Nobody wants to look at a photograph this way. The reason to do it is that a playout server, an IPTV encoder, a set-top ingest, or an HLS-era segmenting tool accepts .ts and nothing else, and you need one Olympus frame to go in as a slate, a test card, or a hold frame. This page covers what actually comes out.

What You Get

How to Convert ORF to TS

1. Upload Your ORF File: Drop the .orf on the page or click "+ Add Files". Queue several frames if you want them in one stream.
2. Set Merge Strategy and Image Duration: Merge strategy opens on Merge images, placing every upload back to back in one .ts; Video per image writes one file each. Image Duration opens at 5 seconds per frame and ranges from 1/60s to 10 seconds.
3. Set Quality and Resolution: File Compression opens on Quality Preset at Very High (Recommended), which does real work here because H.264 is a quality-target codec. Constant Quality gives you the number directly on a 16-to-51 slider that opens at 23. Video resolution opens on Keep original; Fixed Resolutions offers exact frame sizes such as 1920 x 1080 if the receiving system expects a broadcast raster.
4. Set Background Color and Convert: Background Color opens on White and fills any bars left when the frame shape does not match the output shape. Click Convert. Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically after a few hours — no sign-up, no watermark.

Where the Pixels Go, and What the Bytes Cost

Two mechanical facts govern the size of the result, and neither is obvious from the settings panel.

The frame is clamped to 4096 pixels on the long edge. A 20-megapixel OM-D file decoded to 5240 x 3912 in testing, and the frame that reached the encoder was 4092 x 3056 — a 22% linear reduction. Keep original means the clamped size, not the sensor size. Larger sensors lose proportionally more.

Almost the entire file is the first frame. At 1 frame per second, the held duplicates that follow cost essentially nothing. Breaking down a 5-second encode of that clamped Olympus frame packet by packet: the opening key frame was 7,042,173 bytes, and the four frames after it were 8,266, 1,406, 381 and 368 bytes. Extending the hold from 5 seconds to 10 barely moves the file size. Raising the quality, or keeping the resolution high, moves it a great deal.

Measured, 5-second holds of the same clamped 4092 x 3056 frame:

That last row is the lever most people are actually looking for. Dropping the frame to a 1080p-class raster cut the stream by 86% — and if a broadcast chain is going to display it at 1080p anyway, the 12-megapixel frame was never going to be seen.

One detail worth knowing about the quality preset: it is resolution-aware. The recommended tier corresponds to a base quality value of 18 at 1080p, and because a 4092 x 3056 frame carries roughly six times the pixels of 1080p, the pipeline nudges it to 20 — more compression, on the reasoning that artefacts hide better at high pixel density. If you want the 1080p-equivalent setting on a large frame, use Constant Quality and type 18.

Frequently Asked Questions

Why is the .ts silent?

Because there is nothing to make sound from. A still-image source has no audio essence, so the pipeline writes one video stream and suppresses the audio controls entirely — a probe of the output finds exactly one stream. Real broadcast transport streams carry AC-3, MP2 or AAC alongside the video, so if the receiving system requires an audio PID to be present, this file will not satisfy it. Lay a bed under the frame in an editor, or generate a silent audio track and mux it in, before delivery.

Is this the same as an .m2ts from a camcorder?

No, and the difference is measurable in the first byte. This page writes plain transport-stream packets: 188 bytes each, sync byte 0x47 at offset 0. AVCHD's .m2ts prefixes every packet with a four-byte arrival timestamp for 192 bytes total, putting the sync byte at offset 4. Broadcast, IPTV and HLS tooling expects the 188-byte form this page produces. Blu-ray authoring and AVCHD editors expect the other one — ORF to MTS is the route for those.

Which white balance gets baked in?

Your camera's. The raw decoder applies the white-balance multipliers the Olympus body recorded at capture and demosaics to 8-bit sRGB, so the result should look close to the JPEG the camera would have written itself. There is no white-balance control on this page and no auto-correction layered on top. If the shot was under mixed lighting and the body guessed badly, develop the .orf in a raw editor first and convert the exported image instead.

Does the ORF's editing latitude survive?

No. Olympus raw files hold roughly 12 bits per photosite of headroom, which is what lets you recover a blown highlight or shift colour temperature after the fact. The transport stream carries 8-bit 4:2:0 video with the tone curve already applied and colour resolution halved in both directions. Nothing about that is recoverable. Keep the .orf as your master; treat the .ts as a delivery artefact.

Why does the Video Codec list include MPEG-2 and MPEG-1?

Because a transport stream is deliberately codec-agnostic and DVB and ATSC broadcasting ran on MPEG-2 for decades. If you are feeding a traditional playout chain or an older set-top, MPEG-2 may be exactly what it expects. Be aware that switching to it changes how the quality controls behave: MPEG-2 has no quality-target mode in this pipeline, so the Constant Quality value is discarded and replaced by a resolution-derived bitrate. H.264 is the default because it is what modern transport-stream tooling and HLS segments carry.

Can I put several Olympus frames into one transport stream?

Yes. Upload them together and leave Merge strategy on Merge images; each developed frame is held for the Image Duration you chose, in upload order, in one .ts. Video per image produces a separate file per photo. Neither adds transitions, motion or audio — the output is a run of held stills, which is what a slate or test-card sequence is supposed to be.

Why are there white bars around my picture?

Because the frame's aspect ratio does not match the output shape, and the converter pads rather than stretching or cropping. Micro Four Thirds sensors are 4:3, so an Olympus frame put into a 16:9 raster gets bars on the sides as a matter of geometry. Background Color opens on White; pick Black or any other swatch, or set Video resolution to a Width x Height that matches 4:3 and the bars disappear.

Will making the clip longer make the file bigger?

Barely. At 1 frame per second with a motionless picture, the first frame carries almost the whole payload and each held duplicate after it costs a few hundred to a few thousand bytes — measured at 8,266, 1,406, 381 and 368 bytes for the four frames following a 7 MB key frame. Ten seconds costs essentially the same as five. If the file needs to be smaller, change the resolution or the quality value; the duration is not the lever.

Something refuses to play my .ts — is the file broken?

Probably not. Bare transport streams have no index at the front, so consumer players often scan awkwardly or refuse them outright, and browsers will not open one at all. Test in VLC or mpv first, both of which handle plain .ts correctly. If all you actually needed was something watchable and shareable, ORF to MP4 is the right target — and if you only wanted to see the photograph, ORF to JPG gives you the full sensor resolution with no 4096-pixel ceiling in the way.

What happens to my raw file after conversion?

It is uploaded over an encrypted connection, developed and encoded on our servers, and both the uploaded .orf and the .ts it produced are deleted automatically after a few hours. Nothing is shared or made public, no account is needed, and no watermark is applied.