Why it doesn’t makes sense to define 35mm format as standard.

Panoramasicht vom Fellhorn

In the first years of digital photography only sensors, which were smaller in size than 35mm format (24x36mm2), have been available. As the normal user has been used to use the focal length, and the resulting field of view, at this time industry began to specify the focal length of a camera as equivalent to 35mm. This had the advantage that consumers quickly understood which angle of view a lens depicts on the smaller sensor.

When sensors in 35mm format came onto the market, this nomenclature was adopted. This quickly established a test culture in which cameras with a smaller sensor were compared with the 35mm format. For this purpose, the 35mm format was taken as the standard. The recording of the camera with the smaller sensor was made in such a way that it came as close as possible to that of the 35mm camera.

The consequences

This approach led very fast to statements such as: The depth of field of an image depends on the sensor size. What is not correct. Who is interested in further details about this topic should read my blog post “Four influencing image quality factors of digital cameras”.

However, it usually does not stop with the above statements. The testers often go even further and crop all sensors that are used for comparison to the 3:2 format. Why? If you look at the history of photography, you realize that photography has always had many different formats. Nevertheless, you then hear statements like: Despite the same native resolution, the sensor has a lower resolution in the 3:2 format. In my understanding, this is a wrong approach, you could also crop the 3:2 format to the 4:3 format. This would end in a reverse the statement. Here’s an example where I took the Micro Four Thirds camera as a reference and cropped the Sony image.

Samle images taken with the E-M1 Mark III
E-M1 Mark III, 40mm, 1/500s, F2.8, ISO200
Sample image taken with the Sony A7 III
Sony A7 III, 43mm, 1/320s, F2.8, ISO100

Alternatives for the 35mm format as standard

After more than 3 decades of digital photography, it is time to stop comparing with the 35mm format. Instead of doing that we should start to discuss the strength of each format. This would give the end-user the opportunity to select the format on their needs. With a Micro Four Thirds camera, you will never make the same images as with a 35mm camera. Just as you will not take the same pictures with a medium format camera and a 35mm camera. For this reason, it makes little sense to use such comparisons in technology. As users, we have to ask ourselves what is important to us in order to implement our photographic ideas.

In case you have to be as mobile as possible Micro Four Thirds is the camera system of your choice. In case you need to shoot in complete darkness and you need to have the best of the best noise performance then a medium format camera might be the right selection. However at the end of the day, the photographer takes the image. Technology is just a support for doing things easier than without it, it cannot substitute your creativity.

Four influencing image quality factors of digital cameras

Introduction

In marketing, it is popular to have short and easy messages. Those short messages allow reaching a big target group, as nobody has to study the topic in depth. In the case of photography is this message “the bigger the sensor, the better the image quality. At the time the first 35mm sensors, with the size of 36x24mm2, arrived in the market the manufactures called this format full-frame sensor. This name suggests to the end-user that this is the biggest available sensor and a clear statement, that the user buys the best available quality. In this communication, it is not mentioned that the sensor is only one of in total of four components, which are relevant for image quality. But there are more image quality factors of digital cameras.

Beside the sensor three additional components are essentially important for the image quality. Lens, image stabilisation and image processor. How those components influence the image quality I will show with simple examples. I hope my thoughts demonstrate, that in photography sensor size is not the only value which the user should monitor to make the correct decision, if they plan to buy a camera.

The lens as image quality factors

Factors influenced by lens

  • Sharpness
  • Resolution
  • Contrast

The lens is the first element of the image quality chain and therefore the most important. In case your lens is not able to deliver details to the sensor, it is impossible that the result can show the details in the result. Due to the sensor specification, the lens requirements in digital photography are much higher than in analog times. Light rays, for example, should reach the sensor straight. Otherwise light rays which should reach i.e. the red pixel could reach the blue or green pixel. This would cause issues in colour processing, which should be avoided.
In analog time i.e. a 50mm F1.2 lens had 7 lens elements (Minolta Rokkor), today a modern 50mm F1.4 lens (Sony Planar T*FE 50mm F1.4 ZA) has 12 lens elements nearly twice as many. This value shows how big the development in lens technology has been since digital times started. As the lens is the most important element for image quality that some of the photographers invest most of their money in an expensive camera and not in the lens. In case the budget makes this decision necessary personally I would make it vice versa. In the end, you would keep the lens much longer as the camera.

The sensor as image quality factors

Factors influenced by the sensor

  • Sharpness
  • Image noise
  • dynamic range
  • resolution

After the lens, the sensor is the second element in the image quality chain. It has an influence on noise performance, dynamic range, and resolution, which means it influences the sharpness of an image. The sensor size influences the image noise and dynamic range only indirectly, but more in detail now.

Image noise / Dynamic range

If you want to look at the influence of the sensor size on the image noise you first have to make sure that technology improvements do not affect the result. Fortunately, I had a Sony Alpha 6000 and a Sony A7II available. Both cameras were introduced in 2014, so it can be assumed that the sensor manufacturing technology is the same. Also, both cameras have an identical number of pixels with 6000×4000 pixels. To find out how the results differ, I took a picture with both cameras, with the same lens and the same settings. The result looks like this.

Sony Alpha 6000 sample image at 25600 ISO
Alpha 6000, ISO 25600
Sony A7 II sample image at 25000 ISO
A7 II, ISO 25600

Even in the preview, you can see that the A7 II has a larger dynamic range than the Alpha 6000. This is especially noticeable in the mouth. If you look at an enlarged section, it quickly becomes clear that the A7 II is less noisy.

Detail of Sony Alpha 6000 sample image at 25600 ISO
Alpha 6000, ISO 25600
Detail of Sony A7 II sample image at 25600 ISO
A7 II, ISO 25600

It seems that the connection between sensor size and lower noise is confirmed and therefore most of the testers stop at this point with the comparison. But is really the truth? As the A7II as a function with which you can switch the area from 35mm format to APS-C format. In case the sensor size would have an effect this should be visible in images taken in this way. Let’s take a look at the comparison pictures.

Sony A7 II sample image at 25600 ISO in APS-C crop mode
A7 II APS-C setting, ISO 25600
Sony A7 II sample image at 25600 ISO in A35mm format mode
A7 II 35mm format, ISO 25600

As you can see the dynamic range is identical in both images, but what is the image noise.

Sony A7 II sample image at 25600 ISO in APS-C crop mode
A7 II APS-C setting, ISO 25600
Sony A7 II sample image at 25600 ISO in APS-C crop mode
A7 II 35mm format, ISO 25600

Also, the noise is identical. Sensor size could not be the reason, that the noise performance of the Alpha 6000 higher is than the one of the A7 II.
The reason is the pixel size itself. As the Sony Alpha 6000 has the same pixel count on a smaller sensor area, the pixel itself is smaller, or in other words: the pixel density is higher. It isn’t as easy as it looks. The sensor size is no parameter for more or less noise performance, this has to be seen in conjunction with the pixel count.
The pixel size of an APS-C camera with 24 million pixels is equal to the pixel size of a 35mm format camera with 54 million pixels. Would I compare those cameras the noise performance would be identical.