Question for the sensor experts

[GMB]

Advertisement (gone after registration)

Some on this forum have suggested that one reason to keep an M8 in addition to the M9 is the crop factor. Thus, a 90 mm lens would be a 90 mm on the M9 but at 120 mm on the M8. I wonder whether that is really a relevant argument.

 

The FF sensor of the M9 makes full use of image circle of the 90 mm lens. The cropped sensor of the M8 does not; as its cropped sensor suggests, it crops the image at the moment the image is taken and therefore narrows the field to that of a 120 mm lens . On can achieve the same effect (i.e., narrower filed of view) for an image taken with the M9 by cropping the image in post processing.

 

Moreover, the M9 has 80% more pixels than the M8 sensor but the sensor is not 80% larger and the pixel density is therefore higher. Thus, all other things being equal, a M9 image cropped in post processing to achieve the same field of view as a M8 image (taken with the same lens) should show more detail than the M8 picture.

 

This to me would suggest that if I want to achieve the field of view of a 120mm lens, I am better off shooting with a 90 mm on the M9 and crop in post processing than shooting with the 90 mm on the M8 (again, all other things being equal).

 

Correct or am I missing something? I was never good in physics so my thinking may be completely flawed

[georg]

"Moreover, the M9 has 80% more pixels than the M8 sensor but the sensor is not 80% larger and the pixel density is therefore higher"

 

No, the photosite density is the same, so when you crop a M9-image to 10MP you will get a file similar to the M8 (also regarding DoF). It's not a reason to choose the M8.

[GMB]

"Moreover, the M9 has 80% more pixels than the M8 sensor but the sensor is not 80% larger and the pixel density is therefore higher"

 

No, the photosite density is the same, so when you crop a M9-image to 10MP you will get a file similar to the M8 (also regarding DoF). It's not a reason to choose the M8.

 

Sort of what I was saying, except that I think the pixel density on the M9 is not only equal but higher than the M8.

[dpstjp]

Sort of what I was saying, except that I think the pixel density on the M9 is not only equal but higher than the M8.

 

I don't think so. Density (ie pixels per square inch) is the same, the M9's just got a bigger sensor. M8 has 3936x2630 pixels (approx 10MB), and M9 has 5270x3516 pixels (approx 18MB). This gives the crop factor of 1.33x (5270/3936=1.33 roughly) as this is determined by the difference between the lengths of the sensor, and not the areas.

[colorflow]

The pixel density is number of pixels divided by the area of the sensor. The ratio of the M9 vs M8 sensor areas is 1.33x1.33=1.77 and the ratio of their no. of pixels is 18. Practically the same within rounding. So same pixel density.

[GMB]

dpstjp and Alan,

 

Thanks. Back to original question, If I crop a M9 image in post processing, will I get the same result as if I had taken it with a M8?

[luigi bertolotti]

Advertisement (gone after registration)

Yes, with same lens and same distance from subject: this is true in terms of pixel resolution; but M9 has a newer sensor and can process 14 bit per pixel files... easy to think that the final result is someway better than M8.

[lct]

...This gives the crop factor of 1.33x (5270/3936=1.33 roughly) as this is determined by the difference between the lengths of the sensor, and not the areas.

The crop factor is determined by the sensors' diagonal.

http://www.l-camera-forum.com/leica-forum/1023020-post20.html

[colorflow]

Yes, with same lens and same distance from subject: this is true in terms of pixel resolution; but M9 has a newer sensor and can process 14 bit per pixel files... easy to think that the final result is someway better than M8.

 

Also the circuitry around the sensor has been redesigned to minimize noise which appears to be the case looking at the high ISO test shots.

 

However I wonder if the thicker IR filter has any detrimental effect on the image quality?

[luigi bertolotti]

Also the circuitry around the sensor has been redesigned to minimize noise which appears to be the case looking at the high ISO test shots.

 

However I wonder if the thicker IR filter has any detrimental effect on the image quality?[/quote]

Tricky question... but after all also the UVIR permanently in front of M8 lenses should add a marginal degradation...

[t024484]

The crop factor is determined by the sensors' diagonal.

http://www.l-camera-forum.com/leica-forum/1023020-post20.html

 

To calculate the crop factor, you can take width, height or diagonal, they all give you 1.33

 

--------M9 / M8

width 36 / 27 = 1.33

height 24 / 18 = 1.33

diagonal 43 / 32 = 1.33

[lct]

To calculate the crop factor, you can take width, height or diagonal...

Only when sensors have the same proportion, 3:2 for example. Take a 6x6 camera for instance. Would make 0.4 with height or 0.6 with length instead of 0.5 with diagonal.

[ho_co]

t02--

lct is correct. The diagonal is used because it expresses the angle of coverage of a lens, where choosing one side wouldn't. 6x6, 4x5, 8x11 can be directly compared only by considering the diagonal of the format.

 

See the Panasonic and Leica cameras like D-Lux 4, where different parts of the sensor are used for different formats. That is, a "24mm equivalent" will still have the same angle of coverage at 4:3, 3:2 and 16:9 only if the diagonals are equal.