We asked camera companies why their RAW formats are all different and confusing

Original Article ↗ Hacker News Discussion ↗

What “RAW” Actually Is

  • Commenters stress that camera “RAW” is not a uniform or truly raw sensor dump.
  • Files often include on-chip noise reduction, dark-frame subtraction, lens corrections, lossy compression, or even partial HDR/computational photography, especially on phones.
  • A better definition offered: RAW = scene‑referred data (pre–display rendering), not “untouched bits from the sensor.”

Why Proprietary RAW Formats Persist

  • Technically, formats are mostly simple TIFF-like containers with sensor data + metadata; decoding is not the hard part.
  • The real complexity is in interpretation: color science, demosaicing, noise reduction, chromatic aberration correction, AF/WB/exposure metadata, etc.
  • Manufacturers see their processing pipeline and “signature look” as IP and competitive advantage; some treat RAW decoders as trade secrets.
  • Internal toolchains and sensor-tuning workflows are built around proprietary formats; DNG would be an additional format, not a replacement.

DNG: Promise, Pushback, and Patents

  • Many users once standardized on DNG hoping for interoperability, but edits still don’t port cleanly between apps (Lightroom vs Capture One, etc.).
  • Technically DNG is flexible (TIFF-based, extensible tags, can store mosaiced or linear data, supports compression and error correction).
  • Some argue there’s no technical reason cameras couldn’t emit DNG, pointing to Pentax/Leica and Apple ProRAW.
  • Others highlight Adobe’s patent license: compliance requirements, potential IP exposure (e.g., color science methods), and revocable rights make legal departments wary.

Metadata, Sensor Idiosyncrasies, and Experimental Features

  • Extra frames and calibration data (dark/flat frames, sensor profiles, lens-specific corrections, multispectral captures, pixel shift stacks) are often handled in ad‑hoc ways.
  • Open-source libraries sometimes miss or mis-handle this metadata, degrading results versus vendor software.
  • Extensible formats like FITS or generic TIFF could handle such complexity, but either weren’t known or weren’t adopted by camera engineers.

Size, Performance, and Bursts

  • Some users see DNGs (especially linear/debayered ones) as bloated and slow; others show mosaiced, compressed DNGs can match or beat proprietary RAW sizes.
  • Continuous-shooting constraints stem more from sensor readout, buffers, and card bandwidth than from container choice; compressed RAW and fast cards mitigate this.

Impact on Users and Ecosystem

  • Practical pain points: new cameras’ RAW formats lag in third‑party support; some (e.g., Fujifilm lossy RAW) remain poorly supported.
  • Many photographers don’t care about format details as long as their preferred editor supports their camera; perceived lock‑in is limited.
  • Critics argue the lack of open, standardized formats and protocols is part of why the dedicated camera market shrank versus phones and never became a broad computing platform.