A few days ago, I wrote what I thought was a fairly innocuous post about a photographer’s view of the iPhone 5. It was a follow-up of sorts to the post I wrote about how I use the iPad and iPhone in my photography. My thesis was, and remains, that the iPhone 5(s AAPL) camera is a marginal improvement over the iPhone 4S camera, except when using it in low-light situations. I compared it solely to the iPhone 4S simply because I figured that people with an iPhone 4S would be curious how the iPhone 5 camera differed and whether it was worth an upgrade.
Purple haze, all through my lens
When I posted my original review of the camera, I noticed a purple glow in the top left hand corner of the above image. Since there was a blue diode on the speaker and the metal on the desk lamp was blue, I assumed that the purple glow was a trick of light caused by any reflections off the blue materials.
I was wrong.
Instead, the purple haze is a now well-known issue caused by several factors. The most common causes seem to be when shooting into bright daylight or having the light source be just off frame. This report by DP Review gives the best technical analysis of the issue I’ve seen.
Conducting tests, again
Since my initial analysis about the purple haze being caused by an odd light reflection was incorrect, I decided to test out this problem under better control. Previously, I had just tried to get the camera angle as correct as I could. For these tests, I held both iPhones on a stand that was in a consistent spot on my desk (and marked where on my desk it went so I could make sure it didn’t move). Due to the size difference between the two iPhones, the camera angle isn’t exact, but it’s as close as I could get it.
Now, instead of just testing how the iPhone 5 camera handles low-light situations, I was also trying to replicate the purple haze problem.
Test #1: A messy, well-lit desk:
This is shot with my blue desk lamp as the sole light source. My monitor and halogen lamp were switched off.
Not much of a difference, huh? That’s to be expected since at this point it’s not a low-light shot, but a decently lit shot. I think the overall image quality of the iPhone 5 camera is better — mainly around the speaker grill — but I don’t see a big difference.
Test #2: A messy, not-so-well-lit desk:
For this shot, the blue desk lamp was off, the monitor was off. However, I angled the head of the halogen lamp so it shined directly against the wall, indirectly lighting the desk.
For the record: the lighting between these two shots is exactly the same and with no flash. And lookie there, the iPhone 5 shot has some purple in it. What I find interesting is that the main light source is off frame and lighting the scene indirectly by reflecting the light off the wall and it still generated a purple tint. However, in every shot on the DP review, there was some blue in the source material or lighting, I did a follow-up test with just the iPhone 5 camera. In this case, the only light on was the halogen lamp lighting my work laptop off frame. In this case, there’s nothing blue in the scene and I wanted to see how the shot would come out.
So, what does all this really mean?
So far from my tests the purple haze is caused by off-frame lighting, the color blue is present, and the low-light sensor has kicked in. Naturally, you may experience this issue under different conditions. Unfortunately, it’s been raining for two days here so I couldn’t test this under daylight conditions. After spending more time testing and researching this, I still fail to get worked up over it. Yes, in certain conditions you may get a purple haze to your photographs. If this is something that would bother you — and that’s not sarcastic; shooting in those conditions might be standard for you — then the iPhone 5 camera is not the camera for you. That’s OK. There are plenty of cameras out there that will suit your needs.
However, if your aim is to get a usable capture in low-light situations, the iPhone 5 camera is better than the iPhone 4S. Most of the pictures I take with my iPhone are to remember an event by; not create a tack-sharp image.