nikon - Can the sharpness of the lens be evaluated with no relation to the DSLR?

When I read lens reviews, they all say that one lens is sharp, another is very sharp, another one is not sharp at all, etc.

For me, it doesn't make sense to just say that a lens is sharp or not, with no information about the sensor used.

For example:

• 
  
  

  On a 10MP Nikon D60, Nikkor 50mm f/1.8G and Nikkor 18-200mm 3.5-5.6GII are, visually, equally sharp (both at 50mm with the same aperture).

  
  


• 
  

  On a 16MP Nikon D7000, the first lens is ways sharper than the second one.

  
  

Not every reviewer can afford the best full-frame DSLR with the maximum number of megapixels when testing the sharpness of the lenses for the review. On the other hand, rare reviewers who actually have the best camera will write the reviews which may be misleading for most photographers: for example someone who has only Nikon D60 with a Nikkor 18-200mm 3.5-5.6GII and would buy a Nikkor 50mm f/1.8G in order to get sharper images (just for that) will do a mistake.

Can the sharpness of the lens be evaluated with no relation to the DSLR? How the sharpness must be interpreted in the reviews, when the camera model is not mentioned?

Answer

Lenses project a virtual image that has a defined minimum spatial frequency. It does not matter what you use to capture the image the lens is projecting, it can be film, a low res digital sensor, a high res digital sensor, or something that far outresolves the lens itself...that doesn't change how sharp an image the LENS produces, though. This is a bit simplistic, however.

Any real-world system's quality is a factor of its components. A lens has an intrinsic quality in the abstract, as does the sensor, and any other components that may be involved in the process of capturing a photograph (say an extender.) The resolution of the final photograph is a product of ALL the components of the system together, and the lowest common denominator is going to affect the IQ of that image. If the lens outresolves the sensor, the the sensor is probably going to determine the quality of the image. If the sensor outresolves the lens, then the lens is probably going to determine the quality of the image. Of the two are capable of resolving about the same detail, then both factors probably affect quality just as much as the other.

Now, its important to understand the quality of all components individually rather than as a combined whole. Why? Because a single lens may be used on multiple cameras, or a single camera may be used with multiple lenses. The quality of a lens is constant regardless of what camera body it is used on, even if one body has a higher resolution sensor than the other. Similarly, the quality of a camera sensor is the same regardless of what lens its used with. If you only knew the quality of a lens when used on the D7000, you would never know how it might fare when used with a D90, or a D3x. However, knowing the quality of the lens itself without context, you can compute how it might fare on any one of those three cameras.

A key factor in determining the quality of a system is its MTF, or modulation transfer function. An MTF is a way of computing the contrast of a lens or an imaging medium when imaging pairs of white and black lines of a specific thickness, or of progressively finer spacing. The MTF of a lens tells you how much resolution can be resolved starting at its center and progressing through the edge of the lens (most lenses do not resolve perfect detail, and resolve more detail in the center than at the edges.) A useful article that explains resolution, lenses, and sensors can be found at Luminous Landscapes Resolution article. Its rather technical, but explains the situation pretty well. If you are not afraid of math and want to know exactly how MTF describes resolution, you can read Norman Koren's Understanding Image Sharpness (beware, very technical.)