Benchmarking: Nikon D750 vs iPhone 11 Pro Cameras

The iPhone 11 Pro Cameras

The iPhone 11 Pro has three cameras:

Ultra-wide at about 120° diagonal FoV, near 108° horizontal on the 4:3 frame
Wide with ~70-75° HFOV as measured
2X 'Telephoto' with ~35-40° HFOV as measured

D750の設定

I compared the iPhone 11 Pro wide and 2X cameras to my D750 set up with my two prime lenses: the Nikkor AF-S 28mm F/1.8G and Tamron SP 85mm F/1.8 Di VC. The 28mm sits close to the wide camera at 26mm equivalent, the 85mm is longer than the 52mm equivalent 2X, and the ultra-wide near 13mm equivalent has no counterpart in my lens set, so it stays outside the paired comparison.

I originally purchased the D750, because the camera was a user-friendly full format 35mm option under $1500 that utilizes a Sony image sensor. The Sony Alpha mirrorless lineup has better performance (newer sensors, and a shorter flange distance that relaxes the long back focus an SLR mirror box forces on wide-angle designs) but was more expensive in 2017. I selected the lenses as they were the most cost-effective lenses which had acceptable MTF (>20%@ Nyquist) that matches the pixel pitch of the D750.

While the Optical Low Pass Filter (OLPF) on the D750 lowers MTF near Nyquist to suppress aliasing and moiré, I wanted to make sure that my hardware configuration was future-proof. This way the lenses achieve optimal sharpness when I decide to remove the OLPF in the future.

テスト構成

The iPhone pictures were captured in 4k3k JPEG mode (Deep Fusion enabled) and the Nikon pictures were captured in 24MP 6k4k RAW mode. RAW captures minimally processed sensor data with more processing latitude, and can be lossless or lossy compressed, so the results are not 100% comparable.

The measurements were taken using my 54MP-rated 4X Imatest eSFR ISO12233:2017 Chart. I also captured images using my CPIQ Spilled Coins Test Chart.

This was an in-situ test under broad daylight conditions on a sunny day, rather than a laboratory with calibrated illumination.

結果

Resolution: Both full format prime lenses on the D750 measured higher SFR than the two iPhone 11 Pro cameras in this comparison, the wide and the 2X. The D750 records 6016x4016 pixels, about 24MP, so its long side is 6K class and not 4K.

The iPhone 11 Pro has severe texture loss. There is severe texture loss when viewing the CPIQ spilled coins test chart. I did not take the time to run Imatest on these test charts, because the degradation is visibly noticeable.

ベンチマークの主なポイント

Resolution: Both of my full format prime lenses on the D750 measured higher SFR than the wide and 2X cameras on the iPhone 11 Pro. Deep Fusion JPEG output against 24MP RAW is not a like-for-like comparison, so read this as a rank order and not a calibrated margin.

Visible Texture Loss. The iPhone 11 Pro lost coin texture in these daylight frames, and that detail cannot be recovered in post process. The same loss shows up when trees are in the scene. Low light should be worse, but that is an expectation from the noise I saw and not something this test measured.

The Bottom Line: Nikon, Canon, and Sony full format cameras are still demonstrably better than phone cameras! Digital zoom on the iPhone 11 Pro crops and upsamples rather than resolving more detail, so switch cameras or crop later instead. Considering the total track constraints imposed by the cellphone form factor, I'm quite impressed with what Apple has been able to achieve. These cameras are certainly good enough for my daily photography walks. For travel, landscape, and event photography  I'll stick with my Nikon full format DSLR.