summary

Marques Brownlee - What is Happening with iPhone Camera?

Thumbnail of the video “Marques Brownlee - What is Happening with iPhone Camera?”
Community upload. AI output has not been independently verified.

Short Summary

Main recommendation: Judge modern smartphone cameras by their software processing rather than sensor size alone, and expect iPhone photos to keep improving through software tuning after the jump to the 48 megapixel sensor on the iPhone 14 Pro. Marques Brownlee argues the iPhone's recent drop in blind-test results reflects overprocessing and inconsistent handling of skin tones and lighting, not a hardware failure. He still awarded the iPhone 14 Pro the best overall camera trophy because the blind test measured only a narrow set of factors like exposure and color, while sharpness, autofocus, video, and app speed were not tested.

Key Points

  • Smartphone photos are now computational: the phone samples multiple exposures, merges them, and applies tone mapping, noise reduction, and HDR, so software often matters more than the sensor.
  • Google's Pixel 6 shifted to a 50 megapixel sensor and initially overprocessed images, with sharpening and noise reduction piling up; the Pixel 7 with the same sensor returned to the top of the blind test after software adjustments.
  • The iPhone 14 Pro follows a similar pattern: after years of 12 megapixel sensors, its new 48 megapixel sensor produces some photos that look slightly overprocessed.
  • The iPhone tends to evenly light every detected face, which can erase natural lighting direction and produce odd results in low light.
  • Google's Real Tone aims to render a wide range of skin tones accurately, while Apple's approach, as observed, does not account for different white balance and exposure needs for different skin tones.
  • The blind test only judged small, compressed still images, so sharpness, autofocus speed, app responsiveness, video, and audio were not evaluated.

Details Worth Noting

  • Concrete instructions: The transcript gives no step-by-step setup or buying instructions. The only actionable guidance is to judge phones across several factors rather than one test: use a varied photo set (faces with different skin tones, texture, sky for dynamic range, foreground and background blur) and check how each phone handles landscapes, skin tones, and pets, since no single phone wins all of them.
  • The test subjects were the presenter's own face, a figurine, and a sky background, chosen to cover color, blur, dynamic range, and sharpness at once.
  • The Pixel 6a, using the same 12 megapixel-era sensor and tuning Google had used since the Pixel 3, won the December blind test, which suggests tuned software can outweigh a newer sensor.
  • Earlier phones boosted greens for landscape shots and blues for sky shots, showing that phone cameras have long altered reality.
  • Film cameras were historically calibrated for lighter skin, underexposing darker skin tones; the presenter says the RED cameras he prefers render his skin tone accurately.
  • The presenter expects Apple to refine the iPhone 14 Pro's processing through software updates and predicts further improvement by the iPhone 15 Pro.

Bottom Line

Phone camera quality now depends heavily on software choices, so a phone's photos can get better or worse with updates even when the hardware stays the same, and no single phone is best for every kind of shot.

Facts

  1. Smartphone photos are now computational, with phones merging multiple exposures and applying noise reduction and HDR before a photo is saved.

    2:00
  2. Google's Real Tone is a smartphone camera approach that trains software on diverse skin tone data to render faces more accurately.

    13:00