How to convert RAW files to 16-bit TIFF for editing

Published October 23, 2026 · 6 min read

A JPEG stores 256 levels per colour channel. A 16-bit TIFF stores 65,536. Your camera recorded 12 or 14 bits, and a 16-bit TIFF keeps all of them, so you can lift shadows and pull back skies in your editor without banding. It is the file to open in Photoshop, Affinity Photo or GIMP, or to keep as a master.

Measured with Contenta Converter 9.0.43 on one public-domain (CC0) sample file from raw.pixls.us. Both TIFFs use lossless LZW compression. Sizes change with the picture.

RAW to 16-bit TIFF, step by step

1. Open Contenta Converter. On the Files step, click Add Folder or drag the RAW files onto the window. CR2, CR3, NEF, ARW, RAF, ORF, RW2 and DNG are all read.
2. Click Next: Operations. Convert is on. When every file is a RAW, RAW Processing is on too; if not, tick it. Leave Apply Effects off.
3. Click Next: Configure. In the Convert section, choose TIFF.
4. Open RAW Processing. Under Bit depth, tick 16 bits per channel (TIFF). This unticks the Sharpness, Color and Contrast boosts, because they work on 8 bits.
5. Check White balance, and move Exposure if the whole shoot is too dark or too bright.
6. Choose the output folder and click Start Converting.

What keeps 16 bits, and what drops to 8

The 16 bits survive only when nothing in the job works on 8 bits. This is what we measured with version 9.0.43:

A file that cannot be 16-bit is still converted, at 8 bits, and the results screen says why for that file.

Exposure: fix a dark or bright shoot in the RAW

The Exposure slider runs from −3 to +3 EV in steps of 0.1. One stop up doubles the light, one stop down halves it. It works while the RAW is developed, on the full sensor data, which gives cleaner shadows than brightening a finished JPEG.

The same value goes to every file in the batch. Use it for a set with one shared error, such as a whole session shot a stop under. Fix single frames later in your editor: that is what the 16 bits are for.

DNG lens corrections

Many DNG files carry their own lens fixes. Phones, drones and Adobe DNG Converter write distortion, vignetting and shading corrections into the file. Lens corrections from the DNG applies them, and it is on by default.

Files without those corrections are not changed, and the option does nothing to CR3, NEF or ARW files. Untick it if you prefer to correct the lens yourself in your editor.

The camera's embedded preview: fast, but 8-bit

Use the camera's embedded preview converts the JPEG the camera stored inside the RAW instead of developing the sensor data. It is the fastest route and keeps the camera's own look.

But that preview is an 8-bit JPEG, it can be smaller than the full picture, and white balance, denoising and exposure do not apply to it. Use it for quick proofs, not for a 16-bit master. Ticking it unticks 16 bits per channel. If a file holds only a thumbnail, the RAW is developed instead.

From the command line and MCP

The contenta command line does the same. --raw-16bit needs -f tiff and turns the default RAW sharpening off:

contenta batch "D:\Photos\RAW" -o "D:\Photos\TIFF16" -f tiff --raw-16bit --exposure 0.7

Add --lens-corrections false to skip the DNG corrections, or --embedded-preview for fast 8-bit proofs. With --json, every file that could not be written at 16 bits is listed under warnings.

Through the MCP server, the convert_image and batch_convert tools take raw_16bit, exposure, lens_corrections and embedded_preview. More about the CLI and the MCP server

Questions

Related: how to apply a LUT to a whole folder of photos