A RAW file is the raw sensor data from a camera, holding far more information than a normal image and requiring processing before it looks like a photo.[2] A PNG is a lossless image that anyone can open. Converting RAW to PNG develops the sensor data into a viewable picture and stores it without throwing any of that detail away, which is the appeal: you get a shareable file that gives up as little quality as possible.
The price of that fidelity is size. A PNG keeps every pixel exactly, so it lands much larger than a JPG of the same photo. This guide explains what the two formats actually are, how the developing pipeline turns sensor data into an image, precisely what "lossless" does and does not preserve, and when the lossless route is genuinely the right one.
Developing a RAW into a lossless image
Like any RAW conversion, this starts with demosaicing the sensor data into a full-color image and applying a rendering for white balance, contrast, and color. Where PNG differs from JPG is the final step: instead of compressing the picture by discarding detail, PNG stores it losslessly. Nothing the developed image contains is thrown away to save space.
That means the PNG is as faithful a copy of the developed photo as you can get in a common format. Edges stay crisp, gradients stay smooth, and there are no compression artifacts around fine detail. The conversion still applies automatic development choices, so it is not the same as editing the RAW yourself, but the pixels it produces are preserved perfectly.
RAW to PNG develops the sensor data and stores the result losslessly. You keep maximum image quality in a universal format, and you pay for it in file size.
RAW and PNG: two very different files
These two formats are not just different codecs; they represent two different ideas about what an image file is for. A RAW file is not a picture at all in the everyday sense. It is a container, and there is no single RAW standard: each manufacturer defines its own, from Canon's CR3 and Nikon's NEF to Sony's ARW and Fujifilm's RAF.[1] Inside sits the near-linear sensor readout, a small embedded preview image, and metadata describing the exposure and the camera's suggested rendering. The RAW deliberately keeps the creative decisions open, which is the whole reason a photographer shoots it.
PNG, the Portable Network Graphics format, is the opposite: a finished, fully rendered image designed to be displayed. It was created in the mid-1990s as a patent-free successor to GIF and is maintained today by the W3C.[3] A PNG stores its pixels in a chunk-based structure, filters each row of pixels to expose patterns a compressor can exploit, and then packs the whole image losslessly with the DEFLATE algorithm, the same method used inside ZIP archives.[4] Because DEFLATE never discards data, a PNG is a mathematically exact copy of whatever image it was handed, which is precisely why it is the natural target when you want the developed photo kept intact.
How the developing pipeline works
Turning a RAW into any viewable image runs through the same fixed sequence of stages, and only the final packaging differs between a JPG and a PNG. First the converter reads the raw sensor values and applies black-level correction and white balance, scaling the color channels so neutral tones come out neutral. Then it demosaics: because each sensor pixel captured only one color through its color-filter array, the converter interpolates the two missing colors at every pixel from the neighbors, reconstructing a full-color image. Next it applies a tone curve and gamma encoding to lift the flat linear capture into normal contrast, followed by color-space conversion into a standard space such as sRGB so the colors display correctly.
Everything up to this point is identical whether you are heading for a JPG or a PNG. The divergence is the last step. A JPG hands the finished image to a lossy encoder that discards fine detail to shrink the file; a PNG hands the very same image to DEFLATE, which packs it without losing a single pixel. That one difference is the entire reason to choose PNG here: you run the full development, then keep the result exactly.
The demosaicing step deserves a closer look, because it is where much of a RAW's apparent quality actually comes from. A camera sensor does not measure color at each pixel; it is a grid of brightness detectors, each covered by a single red, green, or blue filter in a repeating mosaic known as a color-filter array. Every RAW pixel therefore records just one of the three colors, and the converter must interpolate the other two from neighboring pixels to build a full-color image. A good demosaicing algorithm follows edges and gradients so it reconstructs fine detail cleanly rather than smearing it, and this is precisely the work the small in-camera JPG preview does more crudely. When people say a developed RAW looks sharper and cleaner than the camera's own preview, careful demosaicing is a large part of the reason, and a lossless PNG preserves every bit of that reconstructed detail.
What lossless does and does not preserve
It is worth being precise, because "lossless" is easy to over-read. What PNG preserves perfectly is the developed image: every pixel the converter produced after demosaicing and rendering is stored exactly, with no compression artifacts, no blockiness, and no softening of edges. If you zoom to 100 percent, a PNG shows you exactly what the converter made, whereas a JPG of the same shot would reveal faint block edges and haloing around high-contrast lines.
What PNG does not preserve is the RAW's latitude. The demosaicing and tone decisions were already applied and are now baked into the pixels; the PNG cannot un-apply the white balance or recover a highlight that the rendering clipped, because that information was discarded before the PNG was ever written. In other words, PNG is a perfect record of the developed photo, but the developed photo is already a lossy interpretation of the sensor data. Lossless here means faithful to the render, not faithful to the raw capture.
Convert RAW to PNG, step by step
Open the converter and add your RAW
Open the FileFormer image converter and drop your RAW file in. It runs on your device, so your photo is never uploaded.
Select PNG as the target
Choose PNG from the output formats. The sensor data is developed into a color image and stored without lossy compression.
Expect a large file
PNG discards nothing, so the file will be considerably bigger than a JPG of the same shot. That is the trade for keeping full detail.
Convert and download
Export the PNG and check it at full zoom. It should be free of the compression artifacts you might see in a JPG version.
When PNG beats JPG for a RAW
Choose PNG over JPG when the developed image will be edited further, layered into a composite, or archived and you want no compression artifacts baked in. It is also the better pick when the photo has crisp graphic elements or text overlaid, since JPG can soften those edges. In short, PNG suits any case where fidelity matters more than file size.
If you only need to view or share the photo and size is a concern, RAW to JPG gives a much smaller file that looks nearly identical for casual use.
The bit-depth caveat that catches people out
Here is a detail that surprises photographers who convert to PNG expecting to keep everything. Cameras record RAW at twelve or fourteen bits per color channel, giving thousands of brightness levels, and PNG technically supports up to sixteen bits per channel, so in principle it can carry that extra precision.[4] In practice, most everyday converters and browsers write and display eight-bit PNGs, which hold only 256 levels per channel, exactly like a JPG. That means a typical RAW-to-PNG conversion is lossless in the sense that it introduces no compression artifacts, but it usually still narrows the tonal precision down to eight bits during the render.
The consequence is subtle but real: if you then apply heavy curves to that eight-bit PNG, smooth gradients such as skies can develop faint banding, because there are no longer enough intermediate levels to describe the transition smoothly. This is not a flaw in PNG, it is a limit of eight-bit rendering, and it is the strongest argument for doing serious tonal edits on the RAW itself before you ever export.
The size cost, and keeping the RAW
The large PNG makes this a poor choice for email or fast web pages, where a JPG or WebP would move far quicker. And the PNG, however faithful, still is not the RAW. It does not carry the RAW's wide dynamic range or the freedom to reset white balance, because those live in the sensor data, not in the developed image.
So keep the RAW as your negative for serious editing, and treat the PNG as a high-quality developed copy for wherever a lossless image is needed. The RAW is the equivalent of a film negative: you can produce any number of developed exports from it, but you cannot work backward from an export to the negative. Discard the RAW and you have locked in every rendering decision the converter made on your behalf.
PNG versus TIFF and DNG for a lossless copy
PNG is a fine lossless target, but it is not the only one, and the alternatives matter when the developed copy needs to survive further work.
TIFF is the traditional archival export from a RAW editor, and it has one clear advantage over an ordinary PNG: it routinely stores sixteen bits per channel, preserving the full tonal precision of the render for print and heavy editing without the banding risk that eight-bit output invites. Its downside is that browsers cannot display it, so it is an archive format rather than a web one. PNG hits the sweet spot when you want a lossless developed image that also opens directly in any browser and image tool, which TIFF cannot promise. DNG, Adobe's Digital Negative, is a different animal entirely: it is a standardized RAW container, not a developed image, so converting to DNG keeps the full sensor data and all of the editing latitude while swapping the manufacturer's proprietary format for an open one. Choose PNG for a lossless viewable copy, TIFF for a lossless print or edit master, and DNG when you want to preserve the RAW itself in an open form. If you only need a small shareable file, RAW to JPG remains the lighter option.
Develop your RAW to PNG now
Produce a lossless developed image, all in your browser with nothing uploaded.
Key takeaways
- RAW to PNG develops sensor data and stores the render losslessly with DEFLATE.
- The result keeps full detail with no artifacts, but the file is large.
- Most PNG output is eight-bit, so do heavy tonal edits on the RAW to avoid banding.
- The PNG is not the RAW, so keep the RAW; use TIFF or DNG when you need 16-bit or the sensor data itself.