CR3 to PPM Converter

CR3 to PPM: Canon RAW Into Netpbm's Deliberately Simple Format

PPM is the colour member of the Netpbm family — a format designed in the 1980s to be trivial to read and write rather than efficient. A .ppm is a short ASCII header (P6, the width and height, and a maximum sample value) followed by raw pixel data, with no compression, no colour profile, no alpha channel and no metadata beyond an optional comment line. The Netpbm specification describes the format's own storage as "egregiously inefficient", which is accurate and entirely deliberate.

That makes it a strange destination for a Canon CR3 — a RAW file holding 14-bit sensor data in Canon's crx codec inside an ISO base media container, introduced with the EOS M50 in 2018 and used across the EOS R system since. But it is a common one in image processing, computer vision, scientific imaging and teaching code, where a format you can parse in twenty lines is worth far more than a format that is small. This page renders the RAW frame and writes it as a binary P6 file. There are exactly two controls, and one of them is not what its label suggests.

How to Convert CR3 to PPM

1. Upload Your CR3 Files: Drag one or more .cr3 files onto the page or click "+ Add Files". Batch conversion is supported, and every file runs with the same settings — useful, because a folder of RAW frames is the usual reason to be here.
2. Set Bit Depth: Three entries render — 8-bit (Recommended), which is preselected, 16-bit (High Precision), and 1-bit (Black & White). This decides the maxval written into the header and how many bytes each sample occupies. Read the table below before choosing the third one.
3. Set Image resolution: Image resolution opens on Keep original, which writes the full sensor frame. Resolution Percentage, Preset Resolutions (4320p down to 16P), Width, Height and Width x Height are all available. Because PPM has no compression, resolution is the only thing that meaningfully controls file size.
4. Convert and Download: Click Convert and save the .ppm. Files are uploaded over an encrypted connection, decoded and written on our servers, and deleted automatically after a few hours — no sign-up, no watermark, never shared or made public.

The PPM File This Page Writes

Bit Depth Is the Only Quality Control, and One Setting Is a Trap

Every figure below was measured by putting the same 22.4-megapixel Canon RAW frame (5,796 x 3,870) through the three settings on ImageMagick 7.1.2, the build this pipeline uses.

The arithmetic is worth checking yourself, because in an uncompressed format it always works: 5,796 x 3,870 x 3 bytes is 67,291,560, and the file measured 67,291,577 — the seventeen extra bytes are the header P6\n5796 3870\n255\n. At 16-bit the pixel data doubles to 134,583,120 bytes and the header grows by two characters because 65535 is longer than 255.

The third row is the trap. "1-bit (Black & White)" does not produce a bilevel black-and-white image: the file is still a three-channel P6 with one byte per sample, so each channel is crushed to 0 or 255 and the picture is posterised into a maximum of eight colours — pure black, white, red, green, blue, cyan, magenta and yellow. On a 1,000 x 1,000 crop of our test frame, the 8-bit render contained 66,932 distinct colours and the 1-bit render contained three. And because the specification only switches to two-byte samples above maxval 255, the file is two bytes smaller than the 8-bit version rather than eight times smaller. It saves nothing and destroys the image. Choose it only if hard posterisation is genuinely what you want.

What the RAW Decoder Decides for You

A CR3 does not contain an image in the ordinary sense — it contains sensor readings that have to be demosaiced, white-balanced and gamma-corrected before there are any RGB pixels at all. None of those decisions is exposed on this page, so it is worth knowing which way they go:

• White balance comes from the camera. The decoder applies the white-balance multipliers your camera recorded at the moment of capture, not a fixed daylight matrix. We verified this by rendering a Canon RAW file through the same ImageMagick and LibRaw path this conversion uses and comparing it against two reference renders of the same file: the mean pixel value matched the camera-white-balance render exactly at RGB 118, 91, 63, while a fixed-daylight render of the same frame came out at 119, 69, 21 — a visibly different, much warmer image.
• Exposure, contrast and picture style are the decoder's defaults. There is no exposure slider, no highlight recovery and no picture-style selection here. If a frame is a stop under, it will be a stop under in the PPM.
• The output is sRGB. Wider gamuts such as Adobe RGB or ProPhoto are not offered, and PPM has nowhere to record an ICC profile even if they were.
• The full sensor frame is written. On Keep original you get the decoded frame at its native pixel dimensions, which for a modern Canon body means tens of millions of pixels and, at three or six bytes each, a very large file.

If any of those decisions matter to your result, develop the CR3 in Canon Digital Photo Professional, Camera Raw, Lightroom or darktable first, export a TIFF or PNG, and convert that — the RAW developer is where exposure and colour decisions belong.

Frequently Asked Questions

Why is my PPM more than sixty megabytes?

Because PPM does not compress, at all, in any mode. The file size is width x height x 3 bytes at 8-bit, doubled at 16-bit, plus a header of under twenty bytes — so a 22.4-megapixel frame is about 67 MB at 8-bit and 135 MB at 16-bit no matter what the picture contains. A CR3 of the same frame is typically 25 to 35 MB because Canon's crx codec compresses the packed 14-bit sensor data. The PPM is bigger because it is the fully decoded image rather than a recipe for producing one. The only lever you have is Image resolution: halving both dimensions quarters the file.

Should I use 8-bit or 16-bit?

16-bit if the PPM is going into further processing, 8-bit if it is the end of the line. A CR3 records 14 bits per channel of sensor data; an 8-bit PPM reduces that to 256 levels, which is fine to look at but will band visibly if you later lift shadows, pull down highlights or apply a strong curve. A 16-bit PPM holds the full 14 bits with room to spare, at exactly double the file size and with no other cost — and since most software that reads PPM at all reads both, there is rarely a compatibility reason to avoid it. The one thing to check is that whatever consumes the file handles two-byte, most-significant-byte-first samples.

What does the 1-bit option actually do?

It sets maxval to 1, which crushes each colour channel to two levels and posterises the image into at most eight colours. It does not produce a black-and-white or greyscale file — PPM has no greyscale mode, so the output is still three channels of red, green and blue. It also does not save space: the Netpbm specification stores one byte per sample whenever maxval is below 256, so a 1-bit PPM occupies the same bytes as an 8-bit one. Our measurements put them two bytes apart, and that difference is the shorter header. If you want a genuine bilevel image, that is a PBM, and if you want a small file, choose a compressed format instead.

Is there any way to get a greyscale file out of this?

Not from this page. PPM is defined as a three-channel colour format — its greyscale sibling is PGM and its bilevel sibling is PBM, and neither is offered as a target here. The 1-bit setting does not substitute for either. If a tool has asked you for a greyscale Netpbm file, convert to PPM here and use a Netpbm utility or an image editor to reduce it, or ask whether the tool will accept a greyscale PNG instead.

Which white balance and colour settings are applied to the RAW?

The camera's own white balance, as recorded in the file at the moment of capture, and the decoder's default demosaic, tone and sharpening behaviour. We confirmed the white-balance part directly by rendering a Canon RAW file through the same ImageMagick and LibRaw path this conversion uses and comparing it against a reference render forced to camera white balance and another forced to a fixed daylight setting: the mean pixel value matched the camera-white-balance reference exactly and differed clearly from the daylight one. The output is rendered to sRGB, and PPM has nowhere to store a colour profile in any case. For anything where the colour decisions matter, develop the CR3 in a RAW editor and export from there.

Will EXIF data, the capture date or the lens information survive?

No. PPM has no metadata structure at all beyond an optional # comment line in the header, and the conversion strips metadata explicitly. Camera model, lens, shutter speed, aperture, ISO, GPS position, capture timestamp and the orientation flag are all discarded. That last one is worth noting: if your camera recorded the frame with an orientation tag rather than physically rotating the pixels, there is nowhere in a PPM to carry that information. Keep the CR3 if the shooting data matters, or convert to CR3 to TIFF, which can hold EXIF.

Can I open a PPM in Photoshop or an ordinary image viewer?

Photoshop, GIMP, Affinity Photo, ImageMagick and IrfanView all read PPM directly, as do essentially all image-processing libraries — it is one of the most widely implemented formats in existence precisely because it is so simple. What generally does not read it is a web browser, a phone gallery or a document application, none of which has any reason to. If you need something you can email or preview, CR3 to PNG keeps full colour depth in a format everything opens, and it compresses.

Should I keep the original CR3 after converting?

Yes, without exception. The CR3 is the digital negative: unprocessed sensor data with the full editing latitude intact, from which any number of different renders can be produced. The PPM is one particular render with the white balance, exposure and gamma decisions permanently applied, in a format that cannot even record what those decisions were. If you ever need to re-grade, recover a blown highlight or change the white balance, you will need the CR3. Storage is cheap.

I shoot Canon but my files are .CR2. Does this page take them?

No — this page accepts .cr3 only. CR2 is the older Canon RAW format used by DSLRs up to roughly the EOS 5D Mark IV era, and CR3 replaced it from the EOS M50 onward. The conversion is the same shape and the same controls apply, so use CR2 to PPM for those files. If you are unsure which you have, the file extension is definitive — Canon does not mix them within a single body.