Photography

When does full frame actually matter?

When do the differences between an APS-C and a full frame sensor actually matter in practice?

Asked on Stack Exchange,

Mainly when you want shallow depth of field. For the same subject and framing, full frame gives roughly 1.3 stops less depth of field, and the accepted answer made the gap concrete: matching a 50mm at f/1.4 on the smaller sensor would need a 31mm at f/0.9, which does not exist.

Where the difference is real

  • Shallow depth of field, which is the advantage people are usually buying. One reply working in portraiture called it invaluable and gave the practical version: shoot at f/2.8 on full frame and get that separation sharp, or chase the same look on the smaller sensor at f/1.4, where the lens is noticeably softer.
  • Low light. A larger sensor allows lower pixel density, which one answer connected directly to low-light performance.
  • Wide angles, which are easier to reach on the larger format — though the same answer noted this matters less once lenses designed for the smaller format exist.

Where the smaller sensor wins

The reply that most usefully complicates the picture pointed out that the advantage inverts when you want maximum depth of field rather than minimum. Landscape and anything needing front-to-back sharpness benefits from the crop sensor's deeper field at the same aperture. The framing offered was that full frame is not better, it is different, and that it reads as better mainly because shallow depth of field is fashionable.

The point that gets lost

One answer redirected the question entirely: the two cameras being compared did not only differ by sensor. They were built for different work — one with more precise autofocus and fast continuous shooting for action, the other suited to studio and landscape. Sensor size is one specification among several, and for a lot of photography the autofocus system and frame rate decide more about whether you get the picture.

When it does not matter at all

  • In good light, at sizes people view on a screen, almost any modern camera is sufficient — a point one reply made plainly.
  • The differences widen precisely where the light is difficult, which is the condition worth testing against rather than the specification sheet.
  • The thread is from 2010 and the specific bodies discussed are long gone. The optics have not changed, which is why the depth of field arithmetic still holds.