Sensor size and depth of field

A recurrent question is, "what does a change in sensor size do to the depth of field."

No way to tell.

Now try this: "what does a change in sensor size do to the depth of field, all other factors remaining unchanged."

Getting closer. But we need to decide what "all other factors remaining unchanged" should mean.

Here is a very agricultural set:

• Focus distance the same.

• F-number the same.

• Focal length the same

• COCDL the same.

Then, as we move to the larger sensor configuration, the depth of field does not change.

But that set of conditions isn't likely to well fit an actual scenario.

Here's another set that might be more meaningful:

• Focus distance the same.

• F-number the same.

• Focal lengths to produce the same field of view.

• COCDL the same as a fraction of the sensor diagonal dimension.

Then, as we move to the larger sensor configuration, the depth of field decreases.

Here's another set that might be meaningful:

• Focus distance the same.

• F-number the same.

• Focal lengths to produce the same field of view.

• COCDL the same.

Then, as we move to the larger sensor configuration, the depth of field decreases more than in the first example.

Best regards,

Doug

Doug Kerr said:

A recurrent question is, "what does a change in sensor size do to the depth of field."

Click to expand...

Hi Doug,

That's indeed a frequently recurring question on many fora.

Here's another set that might be more meaningful:

• Focus distance the same.

• F-number the same.

• Focal lengths to produce the same field of view.

• COCDL the same as a fraction of the sensor diagonal dimension.

That might mean:

We use the COCDL as a fraction of sensor size criterion, or

We use the COCDL as a multiple of the sensel pitch criterion, and the larger sensor camera has the same sensel dimensions in its sensor.​

Then, as we move to the larger sensor configuration, the depth of field decreases.

Click to expand...

The added emphasis above is mine.

Indeed, with the required change in focal length to allow a larger image circle making most of the impact. Of course, the larger physical sensor dimensions also require less enlargement to reach the same size output compared to the smaller sensor dimensions. And in general, the larger dimensions usually also come with more sensels of a relatively large size.

It's the difference in sensel size that makes the following scenario the more likely one (I added the word 'NOT the same', which you seem to have left out). There are however also MF digital backs with relatively small sensel pitch, e.g. the 80 megapixel Phase One IQ180 with a 5.2 micron pitch.

Here's another set that might be meaningful:

• Focus distance the same.

• F-number the same.

• Focal lengths to produce the same field of view.

• COCDL the same.

That might mean we use a multiple of the sensel pitch as our COCDL and the sensel pitch is not the same in the two cameras (its sensor has larger sensel dimensions).​

Then, as we move to the larger sensor configuration, the depth of field decreases more than in the first example.

Click to expand...

What remains, is the need to magnify (or not) the image more, or less, for final output, which also means the physical sensel pitch limit on the COCDL comes into play. It therefore remains important to look at the entire chain of components resulting in the final output.

Cheers,
Bart

Hi, Bart,

Bart_van_der_Wolf said:

It's the difference in sensel size that makes the following scenario the more likely one (I added the word 'NOT the same', which you seem to have left out).

Click to expand...

No, that is not what I meant.

The assumed condition was the chosen COCDL was the same between the two "cameras". I said that this might mean that (a) we used as a criterion for choosing a COCDL a fixed multiple of the sensel pitch and (b) (in the case of interest) the sensel pitch was the same, so that we would choose the same COCDL.

I do not claim that this scenario is "likely".

Thanks for your observations.

Best regards,

Doug