How do iPhones produce such good low light images and could mirrorless mimic them?

[Rufus]

Well the title says it all, really!

I am astounded at how well an iPhone copes with low light, and does so far better than my DSLR. How does it do this?

Do you think camera manufacturers will ever catch up with Apple. I would love a mirrorless Canon camera with a full frame or APC sensor that had the same on board processing capability built in to it to provide the same flexibilty as my phone. This would add far more value than some of many the "improvements" manufacturers make to their model line-ups.

[DanK]

They do it with lots and lots of postprocessing and computational stuff.

All DSLRs and MILCs will do some processing for you (shooting JPEG), and have substantial computational photography functions as well, e.g., the OM-1 Mark II. However, the two types of tools are aimed at different markets. To some extent, the DSLR and MILC market is aimed at people who want control. That's true even of some people who shoot JPEG, e.g., sports photographers. Phone cameras are point-and-shoots, designed for people who just want to hold something in front of them that will take a photo. The manufacturers have gotten increasingly clever at using computation to make up for the lack of human judgment in deciding how to capture or process the image.

DSLRs and MILCs, used right, can do low-light photography that phones can't. For example, this photo is a 7-minute exposure, ISO 100, taken around midnight with a Canon 5D Mark III. It's essentially noiseless, and since it didn't need noise reduction, it has a great deal of detail (which you can't really see at this size) that would be lost with noise reduction.

[lunaticitizen]

The iPhone does the processing for you and it is very good at it. When a mighty company employs tens or even hundreds of engineers just for image processing you can expect good things to happen.

[billtils]

Nice photo and nice exposition Dan.

[Rufus]

Thank you Dan for your explanation.

I have since found this article by Robin Whalley on Lenscraft which goes into more detail.

[proseak]

It takes a lot of pics very quickly, aligns them, and merges them.

Computaionally intensive but effective.

[Rufus]

I often find the enforced use of computational photography on iPhones to be a distinct disadvantage. I would rather have an unprocessed file that I can process in my own way and in in my own time. I find I have to undo what the iPhone has done, and in particular reduce the vibrancy and/or saturation.

I will watch with interest to see what Adobe does with its Project Indigo and read some reviews. This year I abandoned Lightroom Classic and moved to Bridge and Photoshop. So I shall need to see if Project Indigo would be applicable outside of the Lightroom ecosystem.

[Rufus]

Success! I have downloaded the Project Indigo app and taken a photo to see if I can get a DNG image onto my Windows 11 PC.

In case it of use to anyone, the process was:

• Take a photo
• Automatically saved to Photos App
• Automatically saved to iCloud Photos
• Open iCloud Photos on the web (not on PC drive)
• Right click on image > choose More Download Options > choose Unmodified Originals > click download icon
• Downloads to an iCloud Photos zip file (in my downloads folder)
• Zip file contains both JPEG and DNG files
• Move files as desired
• Open Bridge and navigate to the DNG file > click on DNG file to open on ACR.

Is there sufficient value in the app for me to keep it if Adobe charge for it in the future? The answer depends on how much that would be. There is plenty of development yet to come, but I would want it to be included as part of the iCloud Photographers plan.