Lossy compressed image format — baked-in camera processing, no grading room. RAW is mandatory for finishing; JPEG only for quick checks.
On set, you quickly notice where JPEG hits its limits: the camera outputs a finished, self-contained image—compressed, with the color space and dynamic range already fixed, with no way back. The format originated in the 1990s as a standard for fast image transfer and is still found in every consumer camera today. Perfect for documentation, set photos, or quick client reviews. For a project where you need to make color adjustments later? A nightmare.
The lossy compression—JPEG uses DCT algorithms and discards information that the human eye supposedly doesn't see—radically reduces file size. A 12-megapixel JPEG file might weigh 3–5 MB. The price: you already get an image with white balance, contrast, saturation, sharpening—all preset by the manufacturer and unchangeable. If you try to change these values later in grading, you quickly lose quality in highlights or shadows because the data space simply isn't there. RAW, on the other hand, stores the raw sensor output and gives you complete freedom in editing.
In practice, we use JPEG files from photo sessions or camera test shots to quickly talk with the gaffer and the production designer on location—"do you see how the light sits in this composition?". Also for archives, when it's just about documentation. Some camera systems save JPEG + RAW in parallel: the JPEG is your quick monitor, your proof for review, the RAW is your actual negative.
For motion capture or video chains, JPEG has long been irrelevant—we work with video codecs like ProRes or DCI there. But with digital photography using stills that will later serve as VFX plates or textures, you need to know: JPEG is just the transport format, not the source. If you want to do color correction in post, you need the original RAW. JPEG is finished—and that's precisely its problem and its advantage at the same time.