Image sensor design has come a long way in just a few years. Some improvements have been slow, quiet and evolutionary. Others have brought a clear step forward in sensor performance. The latest innovation is ‘stacked sensors’, which have brought a big increase in continuous shooting and video performance. So what is a stacked sensor and how does it work?

Let’s start briefly with BSI, or back-side illuminated sensors. In the old days, all the associated circuitry for sensors ran in and around the light-sensitive photosites that actually captured the light values.

Then designers figured out that if you turned the sensor around, so that the circuitry was on the back instead of the front, you could make the photosites bigger and effectively use up the whole surface area of the sensor with no gaps in-between. BSI sensors typically offer better sensitivity, better dynamic range and lower noise as a result.

But where BSI sensors brought improvements in image quality, stacked sensors are designed to offer improvements in speed – specifically the ‘readout speed’ – since this is one of the bottlenecks for camera performance.

A sensor with a slow readout produces more rolling shutter, or ‘Jello effects’ in video, or ‘shutter distortion’ when photographing fast subject movement with today’s high-speed electronic shutters. A faster readout speed can also help with continuous shooting speeds and autofocus performance.