ERF to SWF Converter

ERF to SWF Converter

An .erf is a raw photograph from an Epson R-D1 rangefinder; a .swf is a ShockWave Flash movie. Putting one inside the other only makes sense in one shape: a slideshow. That is what this tool builds — each raw is decoded, held on screen for a chosen number of seconds, and the results are concatenated into a single silent clip. If you have one photograph rather than a set, you get a short movie of that one picture, which is a perfectly valid but slightly odd object. This page covers the timing arithmetic, the controls that actually change the output, and where a Flash file can still be used.

How to Convert ERF to SWF

1. Upload Your ERF Files: Drag the raws onto the page or click "+ Add Files". Order matters — the files are concatenated in upload order. Each raw is decoded and normalised to a PNG before encoding, with the long edge capped at 4096 pixels; an R-D1 file decodes to about 3040 × 2024, so nothing is trimmed. Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically after a few hours — no sign-up, no watermark, never shared or made public.
2. Set the Merge Strategy and Image Duration: Merge strategy opens on "Merge images", which joins everything into one movie; "Video per image" produces a separate .swf for each photograph instead. Image Duration sets how long each picture is held and opens at 5 seconds, with choices from a single frame at 60 fps up to 10 seconds per frame.
3. Choose the Video Codec and Background Color: Video Codec opens on FLV — Sorenson Spark — with MJPEG as the alternative. Background Color opens on White for image sources and fills any letterbox bars created by a fixed frame size that does not match your photographs' proportions.
4. Set the Video Resolution and Convert: "Keep original" is preselected and uses the first image's dimensions. Fixed Resolutions offers exact pairs from 7680 × 4320 down to 256 × 144, Preset Resolutions sets a height and lets the width follow, and Width / Height / Width × Height take explicit numbers. Click Convert to download.

How Long the Movie Will Be, and What It Will Weigh

Still images are encoded at one frame per second, and each photograph becomes a clip whose length is the Image Duration you chose. So the arithmetic is exactly as simple as it looks:

Running time is nearly free in bytes, which surprises people. Because every frame within a held image is identical, Sorenson Spark encodes the repeats as almost nothing. Measured on one decoded R-D1 raw at 3040 × 2024: a 1-second clip came out at 225,422 bytes, a 5-second clip at 242,486 bytes and a 10-second clip at 257,712 bytes. Ten times the running time cost about 14 % more file. What does cost bytes is adding more photographs, because each one is a fresh picture the encoder has never seen.

The Controls That Actually Change the Output

Also worth knowing: the codec and quality settings are taken from the first image in the batch and applied to all of them, so put a representative photograph first rather than an unusually dark or unusually small one.

Flash's Status, and When SWF Still Makes Sense

Adobe ended support for Flash Player on 31 December 2020 and began blocking Flash content from running on 12 January 2021, three and a half years after announcing the timetable in July 2017. No current browser has the plugin and no mainstream operating system ships a runtime.

What that leaves is a file format that desktop media tooling still reads perfectly well — the output above opens in VLC and MPV without complaint, because they use the same open-source demuxers this pipeline does — and that a surprising amount of legacy authoring, kiosk and digital-signage software still accepts as an import format. Those are the real reasons to be on this page. If you want a slideshow that anyone can open, build the same thing as an MP4 with ERF to MP4. If you just want the photographs as pictures rather than as a movie, ERF to PNG gives you lossless stills.

Frequently Asked Questions

Will the slideshow have music or narration?

No. The image-to-video path produces a silent output — a photograph has no soundtrack to carry, and the audio controls that appear on video-to-video pages are suppressed here for that reason. There is no way to attach an audio file during conversion. If you need a soundtrack, assemble the slideshow in a video editor, or convert here and add audio afterwards in an editing tool.

Can I control the order the photographs appear in?

Yes — they are concatenated in the order you upload them, so add the files in the sequence you want. Most file pickers hand over files in filename order, which is why R-D1 files with sequential names usually come out chronologically without any effort. If you need a specific order that filenames do not give you, rename the files with a numeric prefix before uploading.

Can each photograph be shown for a different length of time?

Not on this page — Image Duration is a single setting applied to every static image in the batch. If you need varying timings, the practical approach is to run separate conversions with "Video per image" selected and different durations, then join the clips in a video editor. Note that animated sources are treated differently from stills: an animated image keeps its own native timing and ignores this control entirely, but a raw photograph is never animated, so that does not arise here.

Why is the output so much smaller than the raw files that went into it?

Because almost everything in a raw file is discarded on the way. An .erf is around 10 MB of 12-bit sensor data with a full metadata block; the SWF holds an 8-bit, 4:2:0, lossily compressed picture with no metadata at all. On our test one 10 MB raw became a 242 KB five-second clip. That is the expected ratio for this direction, not a sign that anything went wrong.

Is the picture quality good enough to be worth doing?

For a slideshow, usually. For anything you care about, no. Sorenson Spark is an H.263-derived codec designed around early-2000s web bitrates, and it encodes in 4:2:0 with a coarse quantiser — fine detail and saturated colour edges suffer noticeably compared with a good JPEG from the same raw. Lower the Constant Quality slider toward 1 if you need it sharper, and accept a much larger file.

What happens to portrait-orientation photographs?

They come out on their side. The raw decoder used here does not act on a camera's rotation flag, so a vertical frame is decoded in the sensor's landscape orientation and encoded that way. There is no way to rotate a raw file before conversion either, so a mixed portrait-and-landscape set needs handling upstream: convert the raws to JPEGs, rotate the verticals, and build the slideshow from the corrected images.

Will the R-D1's full 3040 × 2024 resolution be used?

By default, yes — "Keep original" takes the first image's dimensions, and the 4096-pixel long-edge cap for image sources sits well above the R-D1's readout. Whether you want that is another question. A 3040 × 2024 Flash movie is an unusual artefact, and most software that still ingests SWF was written for frames a fraction of that size, so picking a fixed resolution such as 1280 × 720 often produces something more useful.

Which cameras produced the .erf files I am converting?

Epson's rangefinder line and nothing else: the R-D1 from March 2004, which was the first commercial digital rangefinder, the R-D1s from March 2006, and the Japan-only R-D1x and R-D1xG from April 2009. The format is TIFF/EP-based and holds a 6.1-megapixel APS-C CCD's output at 12 bits per sample. If your files came from any other camera they are a different raw format and belong on the tool for that one.