Dedicated Cameras – Part 1

Heads Up!

This article is several years old now, and much has happened since then, so please keep that in mind while reading it.

If you own a dedicated interchangeable Lens Camera (ILC) today, chances are it is a DSLR (Digital Single Lens Reflex). For higher-end ILC cameras, the DSLR has been the defacto standard for almost 20 years. It was made as a digital version of the old analogue film cameras (SLR) before it. A DLSR design is based on a viewfinder that see’s the subject through the lens via a Mirror that is sitting in front of the digital sensor. Once you trigger the release button, the mirror “flaps” out of the way, and the sensor is now exposed to the light through the lens (When the shutter allows). Once the image is taken, the mirror falls back in place and you can once again look through the viewfinder.

Figure 1: DSLR Schematics - Drawing by A. SmallmanFigure 1: DSLR Schematics - Drawing by A. Smallman

 

But things are changing - in the last 5 years another way of designing ILC’s have gained a strong foothold in the marked, and this year has been the major breakthrough as even Canon and Nikon joined the frenzy with a highend Full Frame offering. It’s called a Mirrorless Interchangeable Lens Camera (MILC), and it greatly simplifies the internals of an ILC camera. There is no longer a Mirror to move inside the camera, and the sensor see’s the light at all times (when the shutter allows).

Figure 2: Mirrorless Schematics - Drawing by A. Smallman

 

However - Mirrorless is not new. In fact more than 95% of all digital camera’s today are some variation of a mirrorless camera - think of smartphones, pocket cams, and even most ILC cameras with smaller sensors (like Micro Fourthirds). What’s changing is that Mirrorless is now coming to get the DSLR “large” sensor camera marked.

The Major user difference between DSLR and MILC is:

  • DSLR uses an optical viewfinder through the Lens. - you see the framing and subject through the lens via an optical light path. No power required.
  • MILC uses an Electronic Viewfinder as there is no way of making an optical light path through the lens without a mirror. So the sensor must be turned on, continuously record what is see’s, and that needs to be replayed on a screen or viewfinder for you to use towards composing. Lots of power required :-)

 

The first obvious question is: Why did large sensor cameras still rely on the 50+ year old technology of moving a mirror inside a camera when Mirrorless has been available ever since digital cameras became a reality? The reasons are many, but here are the major ones:

BUSINESS REASONS:

  • It was comparatively easy to simply replace the film with a digital sensor and keep the remaining engineering “the same”. That made it quick to market, fairly easy to make and no change of workflow = More money in the bank.
  • There was a huge collection of Lenses that could still be used on a digital version of the camera house. A few vendors did choose to change their lens mount and give up legacy lens compatibility.
  • The R&D required to solve all the technical reasons below was WAY to great a leap to solve all at once - the technology needed was simply not available at the time and yet the advantages of the digital sensor was SO great that film had to be replaced - otherwise your business was dead.

TECHNICAL REASONS:

  • The very fast Autofocus in DSLR’s is done by a process called Phase Detect, and that requires special phase arrays located in the dedicated autofocus module. These Phase arrays see the world through the lens to focus, and the light gets to the arrays via a mirror - just like the viewfinder. At the time you could not integrate Phase Arrays on the sensor itself, which is needed if there is no mirror.
  • With very small sensors - like in smartphones - the Depth of Field (DOF) is very big and moving the comparably small focus lens requires little energy and precision to acquire focus. Autofocus on small sensors can be done by using software that looks to adjust the focus lens towards getting the most contrast in what’s continuously being recorded from the “always on” sensor. This is a very slow mechanism and it´s useless in low light as there is very little contrast in such a setting. Also, the major flaw is the software does not know whether to focus inwards or outwards in the beginning - so it’s prone to severe hunting. But when most of the scene is always in decent focus (huge DOF), the solution was usable. Full Frame and other “large” sensors have a very narrow DOF at a given aperture and having autofocus done with contrast detect could take ages - not something PRO’s (or most amateurs) can live with.
  • Digital sensors at the time consumed a lot of power and having it always on was impossible once sensors became bigger than a few grains of rice. Especially overheating and battery consumption made it impossible, but also sensor lifetime and how sensors needed a shutter mechanism was contributing. You can read much more about the shutter challenge in large sensor dedicated cameras in this article: The Shutter Conundrum

USER REASONS:

  • Even if an Electronic viewfinder could be made with an always on sensor, the quality and speed of it made it completely useless for PRO’s and amateurs. You would be unable to see enough details, and the lag between recording on the sensor and replaying it on the viewfinder would be intolerable (probably up towards one second at the time).

Fast forward 15 years: Technological improvements and lots of R&D has brought the technology needed to address the above problems. Now mirrorless has moved into the realm of very fast and interactive cameras with large sensors. Today we can:

  • Make electronics that uses so little power it’s feasible to have a large sensor always on along with driving a display for the electronic viewfinder - while still on battery.
  • We can create Phase Arrays that are so small they can be integrated onto the sensor itself - between pixels - without loss of image details and quality. So we use the same tech, it’s just located directly on the image sensor.
  • Our electronics is now fast enough that we can replay what the sensor see’s with so little lag that humans no longer detect it. (< 20ms)
  • We can make screens (big or small) with so many pixels that it BY FAR out resolves the human eye. That means we can display details that even the human eye cannot see.

But why are we interested in Mirrorless as a photographer? DSLR’s work just fine, so you could be persuaded to think this is a marketing stunt from all the camera makers to get us to buy a lot of new gear.... Well, not intirely. The promises and features of current mirrorless end especially those to come, are indeed great, and in Episode two we will explore these features and look at what’s to come.

Tue Wolff Madsen