How Far Have We Come?!

(Apple iPhone 5, ISO 50, f/2.4, 1/3623, a.k.a. All AUTO).
Not bad for a phone camera. I pushed the photo a little bit on purpose to highlight how much data these tiny sensors are able to collect. See more like this at Sorin Vacaru Photography

Sometimes we get pulled into the whirlpool of “newer and better, faster, more and higher” and forget where we started. We forget that a few short years ago, pros were taking amazing pictures with cameras we wouldn’t be caught dead carrying today.

I somehow felt the need to step back and see how far cameras and sensor technology have come over the last ten years.

Here are a few interesting facts based on sensor performance numbers published by DXOmark.com :

1. The top scoring APS-C/DX format camera today, the Pentax K-5 IIs scores higher than ANY camera (regardless of sensor size) prior to July 2008; if we look at full frame/FX format cameras only, the K-5 IIs reigns supreme until December of 2008; this is only FOUR short years ago.
2. The best scoring camera with the much maligned CX (1″) format sensor, the Sony RX100, scores higher than ANY camera prior to September 2004.

What do these two simple facts tell me?

Well, first and foremost, forget about the specs, folks! Camera performance is no longer an obstacle to our creative accomplishments. It ceased to be a while ago.

Second, let’s talk about the future. In the electronics world, there is Moore’s Law. Moore’s Law states that the density of transistors that  can be placed on an integrated circuit doubles every two years. Camera sensors and processing pipelines are not unlike integrated circuits. From looking at historical data spanning over 10 years it appears that sensor quality follows an upward trend where sensors of double the area are ahead by about 4 years.

The CX – DX – FX progression, although not exactly a 1 – 4 – 8 area ratio is close enough. This would mean that in four years top CX based cameras should perform like today’s Pentax K5 and Nikon D7000. In eight years, we may have them perform like a Canon 5D Mark III or Nikon D600. Pretty impressive, don’t you think?!

Before some of you storm the site holding the laws of physics in their right hand, let me say that, although there are limitations to how many photons a small photo site can collect, improvements of the ENTIRE processing pipeline may compensate for those limitations. You can also drop the camera’s full functional specs from your left hand. These ideas refer to the sensor and image processing pipeline ONLY, NOT the entire package (controls, handling, lenses, etc).

I know this is a daring prediction and it is based on incomplete data and science, but in my calendar I marked the first week of 2017 with a note to review and comment on this article. Let’s hope 2017 will find us healthy and happy, each with a great camera in our hands. 🙂