mobile photography technology, culture and community
www.dpreview.com

DxOMark Mobile Report: Apple iPhone 5

Summary

No real surprises in the report from DxOMark's engineers, with the iPhone 5 putting in a decent performance but offering little advantage over its predecessor, reflected in the fact that the new model scored exactly the same as the 4S, with a DxOMark Mobile score of 74.

"The iPhone 5 achieves the same mobile score as the iPhone 4S. A score of 74 is very good, but falls short of the best cellphone camera that we have tested, the Nokia 808 PureView, which scores 81. The iPhone 5's strengths are good color and exposure in most lighting conditions, good detail, even in poor light, and accurate and reliable (if a little slow) autofocus in everyday use".

Where the iPhone 5 isn't so strong is under artificial light, where "it tends to give yellowish results" and when shooting with bright light sources at or near the edges of the frame, where lens flare can be a concern. Another thing that might cause problems, is the fact that "the iPhone 5's live view image only shows 88% of the actual vertical image area, which makes accurate framing rather difficult".

In video mode, DxOMark's engineers reported that the iPhone 5 gives "good detail in outdoor, bright conditions, and offers effective stabilization". As with stills, video footage shows "good, accurate colors, except in low artificial lighting where things go a little yellow". They did find some aspects of the iPhone 5's video mode frustrating, including the fact that "image stabilization can't compensate for walking motion if you're filming while moving, and autofocus is not continuous during shooting". Finally, "..in poor light, video footage from the iPhone 5 is quite noisy, and exposure can become unreliable too".


Still Photography

Color, Exposure and Contrast

The DxOMark team found the iPhone 5 to produce "very pleasant and vivid colors in studio conditions", and that it offered "good exposure for most scene types (above 20 Lux)". The iPhone definitely came out better in bright, natural light than in very dim conditions (especially when shooting under artificial light), offering "accurate white balance on natural images", but showing "slight color shading with some indoor illuminants". Because of this DxOMark scored the iPhone 5 at 84/100 for Color in bright light and 76/100 in low light. Overall though a good performance, with exposure and contast scoring a very respectable 88/100.

Overall DxOMark awarded the iPhone 5, based on both measured and perceptual analysis, scores of

  • 4.5 out of 5 for Exposure,
  • 4.5 out of 5 for White Balance accuracy
  • 4.0 out of 5 for Color shading*
  • 3.0 out of 5 for Color Rendering in low light, and 
  • 4.5 out of 5 for Color Rendering in bright light

*Color Shading is the nasty habit cellphone cameras have of rendering different areas of the frame with different color shifts, resulting in pictures with, for example, pinkish centers and greenish corners. The iPhone 5 didn't exhibit significant shading in anything but the very lowest light conditions (20 Lux).

Noise and Details

DxOMark's engineers reported that the iPhone 5 captures images with "nice details in outdoor pictures", but that in low light conditions the images show "moderate loss of details". In low light they also criticize the fact that "smooth contrast gradations show some staircase effects" and that the noise has "a large grain shape". They noted that Apple has changed the denoising algorithm that controls the compromise between noise and texture, compared to iPhone 4S - see below.

Texture Acutance

Texture Acutance is a way of measuring the ability of a camera to capture images that preserve fine details, particularly the kind of low contrast detail (textures such as fine foliage, hair, fur) that can be blurred away by noise reduction or obliterated by excessive sharpening.

Sharpness is an important part of the quality of an image, but while it is easy to look at an image and decide visually whether it's sharp or not, the objective measurement of sharpness is less straightforward.

An image can be defined as 'sharp' if its edges are sharp and if fine details are visible, but in-camera processing means it's possible to have one of these (sharp edges) but not the other (fine details). Conventional MTF measurements tell us how sharp an edge is, but have drawbacks when it comes to measuring fine detail preservation. Image processing algorithms can detect edges and enhance their sharpness, but they can also find homogeneous areas and smooth them out, to reduce noise.

Texture Acutance, on the other hand, can qualify sharpness in terms of preservation of fine details, without being fooled by edge enhancement algorithms.

A detail of  target made of a 'dead leaves' pattern, designed to measure Texture Acutance. It is obtained by drawing random shapes that occlude each other in the plane, like dead leaves falling from a tree. The statistics of this model follow the distribution of the same statistics in natural images.

In this example from a DSLR without edge enhancement sharpness seems equal on edge and on texture. Many details are visible in the texture.

In this second example edges have been digitally enhanced, and the edge looks over-sharp, with visible processing halos ('ringing'). On the texture part, many details have disappeared.

At first sight, the images from these two cameras may appear equally sharp. A sharpness measurement on edges will indeed confirm this impression, and will even show that the second camera is sharper. But a closer examination of low contrasted textures shows that the first camera has a better preservation of fine details than the second. The purpose of the Texture Acutance measurement is to qualify this difference.

Note: Acutance is a single value metric calculated from a MTF result. Acutance is used to assess the sharpness of an image as viewed by the human visual system, and is dependent on the viewing conditions (size of image, size of screen or print, viewing distance). Only the values of texture acutance are given here. The measurements are expressed as a percentage of the theoretical maximum for the chosen viewing condition. The higher the score, the more details can be seen in an image. 

For all DxOMark Mobile data presented on connect.dpreview.com we're showing only the '8MP equivalent' values, which gives us a level playing field for comparison between phone cameras with different megapixel values by normalizing all to 8MP, suitable for fairly large prints. DxOMark also offer this data for lower resolution use-cases (web and on-screen). For more information on DxOMark's testing methodology and Acutance measurements please visit the website at www.dxomark.com.
The iPhone 5's ability to retain fine texture in images increases noticeably as light levels rise from 20 Lux (3.0 EV: very low light, such as a floodlit building at night) to 700 Lux (outdoors on an overcast day). Note that the results under artificial light are almost identical.
The 100% crops above show clearly how the iPhone 5's noise reduction smears fine detail when light levels drop.
The iPhone 5 obviously uses more noise reduction than its predecessor in low light, meaning fine texture captured is fractionally lower in low light. Once you get into daylight (700 lux and higher) the difference is marginal. In this comparison you can see that the Nokia 808 has a sizeable lead and the Samsung Galaxy SIII has the harshest noise reduction in very low light, but all four phones converge in brighter conditions.

Edge Acutance

Edge acutance is a measure of the sharpness of the edges in images captured by the phone's camera, and again we're only looking at the most demanding of the three viewing conditions that DxOMark reports on, '8MP equivalent'.
Unsurprisingly Edge Acutance is high, even in low light (though it does rise slightly as you move up the brightness scale).
The iPhone 5 does slightly better than its predecessor in lower light, but the main differences you're seeing here are down to edge enhancement processing (sharpening).

Visual Noise

Visual Noise is a value designed to assess the noise in an image as perceived by the human visual system, depending on the viewing condition (size of image, size of screen or print, viewing distance). The measurements have no units and can be simply viewed as a weighted average of noise standard deviation for each channel in the CIE L*a*b* color space. The lower the measurement, the less noise in the image.

 Visual noise is high at very low light levels, but drops to a marginal level once you get out into the daylight. Pretty much all the noise is luminance grain (there's almost no chroma noise, thanks to NR processing).
 In tungsten light there's a little more color noise, but nothing you'd worry about.
 The Nokia 808 is the clear winner here, thanks to its groundbreaking 'PureView' technology. The iPhone 5 is a lot less noisy than its predecessor in low light, but its safe to assume this is down to differences in the level of noise reduction being applied.

Noise & Detail Perceptual scoring

DxOMark engineers don't just point camera phones at charts, they also take and analyse scores of real-world shots and score them accordingly. Their findings for the iPhone 5 were:

Natural scene

  • Texture (bright light): 4.5 out of 5
  • Texture (low light): 3.0 out of 5
  • Noise (bright light): 3.5 out of 5
  • Noise (low light) 2.8 out of 5
Bright light sample shot
 100% crop shows "many fine details"
Low light (20 Lux) studio shot
Large grain noise visible
 Edges are sharp (lady's hair) but some details are lost and large grain noise is still visible in the darker skin areas.

Artifacts

Phone cameras, like entry-level compact cameras, tend to suffer from artifacts such as sharpening halos, color fringing, vignetting (shading) and distortion, which can impact on the visual appeal of the end result. DxOMark engineers measure and analyse a range of artifacts. Their findings after testing the iPhone 5 are shown below:

Pros:

  • No ringing (halos)

Cons 

  • Noticeable flare in some outdoor and indoor pictures. 
  • Noticeable moiré in some pictures. 
  • Loss of sharpness in the corners. 

Perceptual scores

  • Sharpness 3.5 out of 5
  • Color fringing 3.5 out of 5

Measured findings

  • Ringing center: 9.6%
  • Ringing corner 2.1%
  • Max geometric distortion 0.39%
  • Luminance shading 8%

Distortion & Chromatic Aberrations

the graph shows the magnification from center to edge (with the center normalized to 1). The iPhone 5 shows very slight pincushion distortion, with a maximum geometric distortion of 0.39%. You are not going to notice this in normal photography.
Lateral chromatic abberations minimal -  chromatic abberation this small are not seen by most observers.

Autofocus

DxOMark also tests autofocus accuracy and reliability by measuring how much the accutance - sharpness - varies with each shot over a series of 30 exposures (defocusing then using the autofocus for each one). As with other tests these results are dependant on the viewing conditions (a little bit out of focus matters a lot less with a small web image than a full 8MP shot viewed at 100%). Even using the 8MP equiv. condition the iPhone 5 did pretty well, earning an overall score of 74/100 in bright light and 69/100 in low light.

Pros: 

  • Accurate and repeatable autofocus in low light and bright light conditions. 
  • Automatic face detection. 

Cons: 

  • Autofocus is sometimes slow. 
  • Unnecessary refocusing. 
Autofocus repeatability - average acutance difference with best focus: Low light 1.74%, Bright light 4.06%

Flash

DxOMark scored the iPhone 5 70/100 overall for its flash performance, docking points for its lack of power (which means noisy images), autofocus errors (when using flash in very low light) and occasional white balance issues, summarizing the performance as offering "Fair detail preservation but strong residual noise (flash is not powerful enough)"

Pros: 

  • Good white balance. 
  • Good flash uniformity. 

Cons: 

  • A reddish cast is noticeable in low light tungsten plus flash illumination, . 
  • Autofocus is not always accurate when using the flash. 
  • Strong noise (flash is not powerful enough). 

Overall DxOMark Mobile score for Photo: 74


Video Capture

DxOMark engineers put phone cameras through a similarly gruelling set of video tests, and you can read their full findings on the DxOMark website here. We'll simply summarize for you. In bright light DxOMark found the iPhone to produce good results with plenty of detail, with effective stabilization and accurate color. They were less impressed with the lowlight performance when there is a visible "loss of texture", and noted the auto exposure jumping around as lighting conditions changed. They also complained about clipped highlights and poor image stabilization in low light and the fact that there is no continuous autofocus, adding that "Lens breathing is visible and unpleasant" when manual focus is triggered.

Pros: 

  • Nice details in bright light conditions. 
  • Good video stabilization in outdoor conditions. 
  • Good colors. 

Cons: 

  • The stabilization has low performances in indoor conditions for walking movements. 
  • Autofocus only works when manually triggered. Visible focus overshoot. 
  • Strong noise in low light and dark areas with an unpleasant chrominance component. 
  • Exposure oscillations during scene illumination changes and clipped highlights. 
  • Highlights are easily clipped
  • Some strange and unpleasant oscillations in exposure have been noticed.

Overall DxOMark Mobile score for video: 70 / 100

Comments

Total comments: 16
armanius

Maybe I missed it, but there's no discussion about the horrible distortion at the top and bottom edges (in portrait ).

0 upvotes
sjkro

Why are so many pictures up-side-down?

2 upvotes
vsk_suresh

There is a Fix for the Purple Haze, Just rotate the phone upside down and shoot it as normal, the purple haze is gone. :-) I tried it and and wanted to let all of you know...

0 upvotes
cph71

a little tip I read somewhere, is that you can use the control on the headphones to trigger the shoot, thus minimizing camera shake.

1 upvote
keeponkeepingon

"For all DxOMark Mobile data presented on connect.dpreview.com we're showing only the '8MP equivalent' values, which gives us a level playing field for comparison between phone cameras with different megapixel values"

Wow. Just wow.

If you did that for normal cameras imagine the uproar......

Why not just report the best each camera can do?

0 upvotes
reeds

Excellent review. It's great to see serious attention on the cameras that most of us use most of the time.

The Nokia Pureview comparison is fair but for the future I'm more interested in comparisons to phones that are closer in size and weight.

The Nokia 808 is more than twice the volume of the iPhone 5 and about 2 oz/57g heavier than the iPhone 5. (Illustrated here: http://www.sizeall.com/compare/Apple-iPhone-5-Nokia-808-PureView-US-quarter-25-cents-coin/314 )

That's like carrying a second phone and 10 US quarters to get the better camera.

What's the best camera phone in same size and weight class?

0 upvotes
keeponkeepingon

"The Nokia 808 is more than twice the volume". No it's not. Your link does not work but I plugged the two phones in. Height and width are the exact same shape, it's a bit thicker with a bump for the camera sticking out that is a lot thicker. To get twice the volume you must be calculating the volume of a box it would fit in. But it's not a box, It's like a slightly thicker iphone with a bump sticking out and given the variation in phone shapes I would consider it in the same class.

1 upvote
reeds

You're right, my first rough estimate was overzealous. Here are the exact values:

The Nokia 808 is 73.2% larger and 50.8% heavier than an Apple iPhone 5 based on both company's published specs. (The 808 is 95.50cc vs iPhone 5's 55.12cc. The 808 is 169g vs iPhone 5's 112g.)

That seems like a big difference to me based on my experience with other devices. However, I have not held them in my hands so I value your opinion. Do you think these specs suggest they are in a somewhat different category? That is, the 808 comparison is interesting but other devices closer in size/weight should be included in the future?

Comment edited 4 minutes after posting
0 upvotes
Hector1980

I've always been a big fan of dpreview, so it's great that now you decided to start testing smartphones.

Are you going to test apps, too? I'm a big fan of this one:

http://jag.gr/645pro/

And it would be great to have a dpreview-caliber review about it.

0 upvotes
vlad0

I've been taking pics with my iPhone 5 for the past week.. it does very well for it's sensor size, but when I put it up against my Nokia 808.. things don't look so good, especially at night..

Here is my little test from couple of days ago:

skydrive gallery: http://sdrv.ms/QOzrHB

4 upvotes
Reg Natarajan

Great gallery. Nothing beats seeing back to back comparison results, and you're right, the Nokia is vastly better.

0 upvotes
cluening

Thanks for the samples. The 808 sure bests the other 2 cams. But the N8 also does quite well, it can be PP'd still very well. Just the contrasts are a bit low, thats it.

0 upvotes
reeds

Great gallery! The last two images really drive home the difference.

1 upvote
Biggus

hmm, I don't see that the iPhone 5 image is that much better than the One X or the S3. It may just about have slightly better resolution, but not in the parts where it's overly bright in the highlights in comparison, indeed in some places, it's the worst of the 4. I'd also say the colour is as much on the warm side as the others are on the cool.

Looks to be a rather biased review tbh.

In my own order of preference, looking at the overall image, I'd choose in the following order...

S3 - most balanced, but less cool than the iPhone is warm
iPhone5 - too warm
808 - soft/suffering from haze
One X - very warm, too dark in the shadows

0 upvotes
Barney Britton

We also incorporate DXO's test data when drawing our conclusions. And we'll have an S3 review online very soon so you can take a more detailed look at how both cameras compare.

1 upvote
cluening

What you consider as haze and softness on the 808 is just the very defensive approach on in-camera tweaking. As I found out with the N8, this haze and softness can easily give perfect pictures in PP.
A lot of headroom.
Cant do with Apple phone cams, as I tried.
Might be the same with the S3, already showing lots of sharpening artifacts, so dead end.

1 upvote
Total comments: 16
About us
Sitemap
Connect