I would agree that the physical aperture of a lens does not change. The challenge comes when considering the equivalent values to compare the performance of a lens on a crop-sensor camera, as opposed to that of the same lens on a full-frame camera under the same circumstances.
I will stress that the physical focal length and aperture of a lens do NOT change. That is not the issue, it's about the interaction of the lens and sensor when producing a result .
To quote the article from DPReview by Richard Butler:
“You do not have to multiply the f-number by the crop factor, unless you want to understand its behaviour, relative to another system, considering the combined effect of the aperture and sensor size."
In the same way that Equivalent focal lengths describe the effect of sensor size on the field-of-view a lens gives, Equivalent apertures describe the effect of sensor size on the properties that aperture affects (depth of field, diffraction, total light projected). In both instances, the underlying properties are not changed: neither the focal length or F-number of a lens is changed by different sensor sizes, only their effects.
- f/number = focal length/aperture diameter
- Equivalent f/number = equivalent focal length/aperture diameter
Comparing equivalent apertures allows you to understand how much control a lens will give you over depth-of-field. It also gives a good idea of how low-light performance will compare between two cameras of different sensor sizes, since it tells you how much total light is making up the final image (most noise in most images comes from the amount of light captured). "
This is previously explored by Richard Butler in his DPReview article: What is equivalence and why should I care? Published Jul 7, 2014 – To quote:
“Conveniently, it's become fashionable for manufacturers to produce 85mm equivalent, F1.2 lenses, which makes it relatively easy to demonstrate the ways in which they are, and aren't equivalent.
In a You Tube video, Tony Northrop demonstrated this effectively - although I disagree with his presentation, as he falls into the trap of poor terminology in that the focal lengths and apertures do not change, but to achieve the same Depth of Field, the crop sensor does have an impact: https://www.youtube.com/watch?v=f5zN6NVx-hY
This is firmly upheld by the Gaussian Lens Formula, as explained by Prof Marc LeVoy, Emeritus Prof in optics and computational photography at Stanford:
The dispute has always been a bugbear because folks don't separate the difference between physical characteristics of the gear - in this case focal length and aperture, as opposed to their effect when combined with a sensor that crops the resultant projected image.
In effect, each camp is correct within limitations. Focal length and aperture are physical characteristics and as such stay constant within the gear. It is the result of combining those characteristics with a different size sensor that creates differences in Field of View and Depth of Field, both of which are heavily impacted by sensor size.
For a full explanation of this see: Equivalence.
Really, this is because the terminology is poor, confusing and misused: resulting in a needless and apparently endless debate - we are essentially arguing 'apples and oranges'.
