Background blur and its relationship to sensor size
1 Background blur and its relationship to sensor siz
Most experienced photographers know that they can modify the amount of background blur by adjusting the aperture of a camera. A typical use case is a portrait of a person, in which the person is in focus and the background is blurred in order to visually isolate the person from the background. But also the sensor size plays a critical role in determining how much background blur the photo will have at a certain aperture and focal length setting. Today's market offers a wide variety of cameras, using all sorts of different sensor sizes, maximum aperture setting and zoom ranges, making it virtually impossible for the user to estimate how much they will be able to adjust the background blur with a certain camera. This article tries to provide a practical guide on how to estimate the maximum strength of background blur for a given camera / sensor type.
In order to be able to say whether a camera is capable to create background blur, we first need to define the term "strength of the background blur". The Depth of field and Lens (optics) Wikipedia articles are good references for the basic formulas that I will use.
Most explanations about background blur start with the term Depth of Field (DoF). While Depth of Field is good, when defining the area of subject sharpness, it can be somewhat misleading if you try to translate the characteristics to the strength of background blur. If you e.g. caluculate the Depth of Field with a DoF calulator first for a subject distance of 1m using a 50mm focal length and then another time moving to a distance of 2m and increasing the focal length to 100mm to reach the same subject framing, you will notice, that the Depth of Field is the same in both situations. This is however not true for the amount of backgrund blur. The further you move away from the subject and compensate this by a stronger zoom, the stronger the background blur will get. The reason for this is, that the stronger zoom results in a narrower angle of view, which will capture less of the background compared to the shorter focal length. This smaller part of the background is however "stretched" to the same frame size, leading to a larger amount of background blur. In the example above moving from 50mm to 100mm will roughly double the strength of background blur, as you can see from the calculations below. So Depth of Field is not the same as strength of background blur.
How strong the background of a photo is defocused can best be seen when taking a look at photos that have spot highlights in the back. Those spots will appear as “blurred discs” in the shape of the aperture when getting defocused, as can be seen in the picture below.
|Photo with strong background blur, with clearly visible blur discs in the back|
The blur disc size b is approximately equal to (please refer to the DoF Wikipedia article):
b = f ·ms / N = (f ·w2) / (N ·w1),
- f is the focal length (the actual focal length, not the 35mm equivalent),
- N is the aperture or f-number,
- w2 is the sensor width,
- w1 is the subject height in portrait orientation or picture width in landscape orientation and
- ms = w2/w1 is the subject magnification of the lens.
Please note, that the equation above is only valid, if the background is sufficiently distant from the subject.
The absolute value of b does not tell much by itself, when comparing cameras with different sensor sizes. If we however compare b to the sensor size w2 we can estimate how big the blur disc size is compared to the total picture height (please refer to the photo above) and thus come to a definition of the strength of the background blur B:
B = b / w2
Example: if the blur disc is 1/1000th (i.e. 0.1%) of the sensor width, the background is almost as sharp as the focused area. If on the other hand the blur disc is only 1/10th (i.e. 10%) of the sensor width, then the background is significantly blurred, as the blur disc covers 10% of the entire picture.
Merging the formulas above yields the following equation for the strength of the background blur
B = f / (N ·w1)
From this equation, you can already see the basic dependencies. The background blur increases
- the more you zoom (i.e. increase the focal length f),
- the more you open the aperture (i.e. smaller f-numbers) and
- the smaller the subject is.
For a typical portrait of a person the photo should probably cover the head, the shoulders and a part of the upper body. A rough estimation is, that the value will be around w1 = 0.6m. This will of course depend on the size of the person and obviously on how much of the person you want to cover in the portrait.
As you can see, estimating the maximum strength of background blur can easily be done by using data that is right away available in the camera specifications. In fact you can even estimate it from data that is available on the lens. Just take the maximum zoom (the actual value, not the 35mm equivalent), divide it by the maximum aperture for the maximum zoom, and then divide this by a typical portrait height of w1 = 0.6m. In the next section I will explain how to interpret the resulting value for B.
Examples pictures that help to understand the values of B
The photos below were taken with a Nikon D7000 using a 18-200mm lens using a variety of aperture and focal length settings in order to be able to create a wide variety of background blur values (please have a look at the next page to see more steps). Please note that although I used the focal length setting equal to 200mm on the lens itself, the actual focal length for a distance of s1 = 2,2m is rather around 145mm. My assumption is that there is a variation in the field of view due to the internal focusing mechanics of the lens and that the actual focal length varies with the subject distance. I tested this and a focal length of 200mm can only be reached for subject distance well beyond 10m. This effect needs to be taken into account when calculating the maximum blur.
In order to categorize the strength of the background blur I used the following definition for B
|0,00% < 0,375%||No blur|
|0,375% < 0,75%||Very limited blur|
|0,75% < 1,50%||Limited blur|
|1,50% < 3,00%||Reasonable blur|
|3,00% < 6,00%||Strong blur|
|> 6,00%||Very strong blur|
Of course there is no such thing as a fixed definition for the strength of background blur and your taste may vary. It's also clear that a camera that is capable to produce B = 1.4 is not much different from a camera with B = 1.6. The idea is just to give an approximate sense to the numerical value.
In order to be able to compare various cameras, we need to define a common scenario. As said above, the smaller the subject the easier it is to create decent background blur. So it does not make sense to compare photos of very small subjects, as basically all cameras will be able to create some kind of background blur. If on the other hand the subject is very far away, you are most likely taking a picture of a landscape for which the background is supposed to be sharp all the way through. Hence the most interesting scenario is to evaluate the before mentioned portrait situation with w1 = 0.6m (head, shoulders and a part of the upper body). I will compare the cameras in two different ways:
- using the maximum zoom, resulting in a maximum achievable background blur for a given camera.
- using a 85mm equivalent focal length setting (which most cameras and standard kit lenses are able to achieve), so that all photo will have the same field of view (subject distance s1 = 1.5m for w1 = 0.6m)
Comparison using maximum zoom
The table below shows a comparison of the background blur B utilizing the maximum zoom available on the respective camera.
|Comparison of the maximum achievable background blur B for a portrait height of w1 = 0.6m|
Comparison for 85mm equivalent
The table below shows a comparison of the background blur B utilizing the 85mm equivalent zoom.
Discussion of the results
Limited background blur
- As expected the iPhone (as well as every other phone on the market) is not capable to produce any kind of background blur for portrait photos. The subjects will have to be fairly small and you have to move very close in order to be able to finally get some kind of background blur (see. e.g. iPhoto 5 example)
- Normal compact cameras are only slightly better than phone cameras and in practice they will not be able to produce significant background blur.
- Cameras using a 1/7" sensor like the Canon S110 and Canon G12, which also have a slow lens at telephoto are also very much limited in their ability to create background blur. This might come as a surprise to some people, as the cameras are clearly targeted at the more serious photographers having lots of manual controls. Those manual controls are however close to worthless, when it comes to taking portrait photos.
- The Nikon 1 cameras combined with their standard lens, does not look very promising either. The small zoom range and the slow lens seriously limit the ability to create background blur.
Cameras which at least provide some capabilities to create decent background blur
- Cameras like the Panasonic LX7 or the Olympus XZ-2 (and XZ-1), are better designed as they make use of faster lenses at telephoto.
- The Sony RX100 unfortunately lacks a faster lens, but compensates this with a larger sensor providing similar capabilities like the LX7 and XZ-2.
- The Canon G15 now got a faster lens compared to the G12. With respect to background blur it is not much different from the Canon G1 X, which has a larger sensor, but a slower lens. The Nikon P7700 is very similar to the Canon G15, having a slower lens but also a longer zoom range.
- FourThirds cameras using standard kit lenses are somewhat limited in their ability to create background blur, but the system offers a variety of faster lenses and longer zooms, which improves the situation signifcantly (like e.g. the 45mm 1:1.8)
- Cameras using APS-C sized sensors are a bit better than FourThirds, but also here the standard kit lenses are a bit short on the telephoto side and too slow. Taking the next bigger zoom range is usually a better option, providing much more photographic freedom.
- And finally it's clear that a 35mm full frame camera like the Canon EOS 6D, combined with a fast portrait lens, provides all the freedom you want. Even the Sony DSC-RX1 that uses a 35mm focal length can produce background blur.
As shown above, estimating the maximum strength of background blur is very easy. Just take the maximum zoom (the real value, not the 35mm equivalent), divide it by the maximum aperture at maximum zoom and then divide this by a typical size of the subject you want to take photo of (as said above 0.6m - 0.7m are good reference points for portraits). On the next page you will find additional pictures, that may help you to find define your own reference point.
|The Lone Photographer by ed rader|
from My Best Photo of the Week
|Neighbourhood Watch by Stevie Boy Blue|
from Zoo trip ~ Cute...
Sony has taken the wraps off of its new 24mm F1.4 GM full-frame lens, which the company claims is the lightest in its class. Despite its fast aperture, the 24mm F1.4 is remarkably light, weighing just 445 grams (15.7 ounces). The lens will set you back $1400 when it ships next month.
In this episode of DPReview TV we take a look at Sony's brand new 24mm F1.4 GM lens, a desirable focal length for many photographers. How does it perform? Chris and Jordan give us their first impressions.
We've had a little time to shoot with Sony's new wide/fast prime, both close to home and on the water in San Francisco. Check out our initial sample images.
Fujifilm released a firmware upgrade for its X-T3 mirrorless camera that addresses issues with distortion compensation and the mechanical lock on SD cards.
The app's algorithms have been trained using using 200 million cropping data points from real photographers.
Thanks to a software update, the Loupedeck+ editing console can now be used for video editing.
British photographic engineer MTF Services is claiming the world’s first third-party lens adapters for the new Nikon Z system with a collection of four units designed to allow cinema lenses to be mounted on the mirrorless full frame bodies.
Think Tank Photo has updated its line of heavy-duty rain covers and introduced a new, compact version for emergency situations.
The X-T3 is our first opportunity to analyze what's likely to be Fujifilm's next generation image sensor. Take a look at how it performs next to the competition in our studio test scene.
Canon's new normal is seriously sharp wide open. After shooting with it for a few days, we've prepared a gallery of real-world sample images.
Nikon will cease offering Brazil-based customer service and technical support, though the company stresses that it will still offer technical assistance and warranty repairs for valid warranties.
Two years ago, CatLABS of JP announced a plan to save Packfilm from the dead. Now, it's announced it's giving up its efforts to better focus its resources elsewhere.
The GoPro Fusion is designed to make it easy to capture 360-degree video and stills. We took it out recently on a typically hot Seattle summer day to see what it can do.
We've got our hands on a full-production Nikon Z7 camera and have updated our gallery with additional samples.
A new Kickstarter campaign seeks funding for Chroma Chrono, a programmable RGB camera flash that emits multiple colors during long exposures.
Think Tank Photo has launched a new lineup of six dual-access, water-resistant protective lens cases it calls Lens Case Duo.
Canon and Nikon finally entered the full-frame mirrorless market this summer with the brand-new RF and Z mounts. Now that we've had some time with the cameras, we wanted to revisit our earlier predictions and take stock.
The devices' camera specs look pretty much identical to last year's iPhone X but under the hood a number of important improvements have been made.
Blackmagic Design has announced the public beta of its new Blackmagic RAW video codec. The company says the new format combines the benefits of shooting Raw video with the ease of use and smaller file sizes usually associated with non-Raw video files.
Serif, the company behind the Affinity suite, has announced the latest update for its mobile Photoshop competitor Affinity Photo for iPad.
The Atomos Ninja V external video recorder and monitor will be ready to ship at the end of this month. The 5.2in Ninja V is designed to provide a smaller option, while still offering many of the features of the larger 7-inch models.
Having shot with the camera, spoken to Canon and read the tea leaves, here's what DPR Technical Editor Richard Butler thinks the EOS R tells us about Canon and the RF's mount's future.
After last week's teaser, lighting manufacturer Profoto has announced its 'small big' new product. The B10 is designed to be used as studio flash head but in a very small body, and has a powerful continuous light source for videographers as well.
Konseen has launched Photo Studio, a new light box tent large enough to photograph people, as well as objects.
Seagate has introduced new high-capacity hard drives for Network Attached Storage (NAS) devices: the 14TB IronWolf and 14TB IronWolf Pro HDDs.
The case was first announced earlier this year as a Kickstarter campaign and comes with a range of features aimed at iPhone photographers.
Manfrotto has introduced a new two-in-one tripod to its Befree lineup. Called the Befree 2N1, this new addition is both a tripod and monopod in one and is available with both of Manfrotto's locking mechanisms.
This new high dynamic range editing software comes with an AI-powered Quantum HDR Engine for improved photo merging.
Apple has unveiled the next generation of its iPhone X in the form of three variants: the 5.8" iPhone XS and 6.5" iPhone XS Max with OLED screens, and the 6.1" iPhone XR with an LCD and single rear camera.