With the release of iOS 27, Apple is introducing a new version of its RAW , revolutionizing the way users edit and manage their photos. This new technology uses machine learning to significantly improve detail and reduce noise in images, including the ability to reprocess older RAW photos, bringing a new dimension to photo editing.
See also: Apple: All the new improvements in iOS 27 Beta 3
The RAW format is a favorite among professional photographers and photo enthusiasts alike, as it preserves the data captured directly from the camera's sensor. This allows for greater flexibility when editing elements such as exposure, color, and white balance. This capability allows photographers to take full advantage of their camera's capabilities, achieving the desired result with greater precision.
With the introduction of RAW 9 , Apple is upgrading its existing line of third-party camera RAW processing, accessible in apps through Core Image . This line includes support and calibrations specific to nearly 800 camera models, making it one of the most comprehensive and flexible solutions on the market.
Apple's history with RAW processing is long and successful. Over the years, the company has updated its RAW processing algorithm eight times, improving the way it handles sensor data, demosaicing, noise reduction, and adjustments like white balance, exposure, color, and tone. These improvements have made RAW processing more accessible and efficient for Apple users.
The new version, RAW 9, is described by Apple as “the biggest update to date.” David Hayward, Core Image Engineer at Apple, said that RAW 9 dramatically improves the performance of RAW files. It is built on a tiled CoreML model that combines demosaicing with denoising for the best quality, and the model runs on the device using Apple Neural Engine for optimal performance.
See also: iOS 27: All the new features of Apple Maps
Hayward showed several examples of RAW 9 in action, comparing the results to RAW 8 and, occasionally, to the original, unprocessed sensor data. In one of these examples, a zoom-in of a low-noise image from a Sony Alpha 7 II shows an old marker. While the image looks good with RAW 8, under RAW 9, the image is sharper, cleaner, and fine text is more legible.
The differences are even more dramatic with high-noise images. For example, a very noisy ISO 51,200 from a Canon 5D Mark III shows a 10x magnification of a box of crayons. The RAW data contains significant luminance and color noise, making it difficult to distinguish the unique color of each crayon. RAW 8 did an acceptable job of recovering true colors, but RAW 9 significantly improves the result, making colors accurate and well-defined, with bright glossy spots visible.
Another example involves a close-up of a photo of embroidery threads taken with a Fujifilm X-T5 at ISO 12,800. This camera has a non-traditional sensor pattern, which is difficult to demosaicing. The RAW 8 results show some color artifacts and loss of detail in the threads, but under RAW 9, the results are noticeably better, with more legible small text and clearer texture in the threads.
For developers, the session describes how to enable RAW 9, optimize performance for editing and exporting, and more. This feature opens up new avenues for developing applications that take advantage of RAW 9 capabilities, giving users even more powerful tools for editing their photos.
See also: iOS 26.5.2, Safari 26.5.2: Apple fixes serious vulnerabilities
The introduction of RAW 9 in iOS 27 is not just a technical upgrade, but a significant development that affects both professional and amateur photographers. The ability to reprocess older RAW photos with this new technology means users can rediscover their old photos with new eyes, taking advantage of the latest advances in image processing.
