12MP, 16MP or 24MP for the 17-55 lens?

Started Apr 19, 2013 | Questions thread
MiraShootsNikon Senior Member • Posts: 1,039
Re: Maybe a different question?

photoreddi wrote:

MarkJH wrote:



If you want to consider 1:1 views on your monitor a measure of "sharpness" or other kinds of image quality, then I suppose this question has an answer.  At some point moving up the megapixel count, those pixels will get smaller than the smallest circle of confusion the lens can project, and at that point a 1:1 view will look blurry.

You're confusing "circle of confusion" (a term related to a lens's depth of field) and "Airy disc", which is the smallest image that a lens can produce from an infinitely small (theoretically) point sized object.

You're confusing "Airy Disc" with "Circle of Least Confusion," which is, without a doubt, what MarkJH meant to say, here.

Let's talk about Circle of Confusion--or at least define it.   Wikipedia does it better than I can:

  • In optics, a circle of confusion is an optical spot caused by a cone of light rays from a lens not coming to a perfect focus when imaging a point source. It is also known as disk of confusion, circle of indistinctness, blur circle, or blur spot.
  • In photography, the circle of confusion (“CoC”) is used to determine the depth of field, the part of an image that is acceptably sharp. A standard value of CoC is often associated with eachimage format, but the most appropriate value depends on visual acuity, viewing conditions, and the amount of enlargement. Properly, this is the maximum permissible circle of confusion, thecircle of confusion diameter limit, or the circle of confusion criterion, but is often informally called simply the circle of confusion.
  • Real lenses do not focus all rays perfectly, so that even at best focus, a point is imaged as a spot rather than a point. The smallest such spot that a lens can produce is often referred to as the circle of least confusion.

Wikipedia goes on to point out how the concept of "Circle of Confusion" might be used

  • For describing the blur spot achieved by a lens, at its best focus or more generally. Recognizing that real lenses do not focus all rays perfectly under even the best conditions, the termcircle of least confusion is often used for the smallest blur spot a lens can make (Ray 200289), for example by picking a best focus position that makes a good compromise between the varying effective focal lengths of different lens zones due to spherical or other aberrations

By these definitions, anyway, MarkJH's use of the term "Circle of Confusion" was absolutely, technically correct in every sense.

Moreover, though I'll let him speak for himself, I suspect he stayed away from talking about "Airy Discs" because he didn't want to get into the effect of diffraction patterns across pixel spreads for such a simple explanation.  Hey, for all your "comic-book-guy" smarts, you didn't, either!

The problem with talking about Airy Discs is that they're a theoretical, not a practical concept.  Let us look at a good working definition--again, I defer to the wikipedia:

  • In optics, the Airy disk (or Airy disc) and Airy pattern are descriptions of the best focused spot of light that a perfect lens with a circular aperturecan make, limited by the diffraction of light.

So If I were wondering: what is the diffraction-limited resolution of a given sensor?  Then I would turn to the concept of the Airy Disc to provide a theoretical limit answer. However, the OP isn't wondering about that, at all.  He's wondering about whether his lens's projection will look "sharp" on his camera's sensor over a variety of commonly used apertures and circumstances. My point, here, is that we're not talking about theoretical limits.  We're talking about a specific lens, with specific optical properties.  (Spherical aberrations, non-circular aperture, etc.)  Which makes Circle of Least Confusion the much more appropriate way to talk about what the OP might expect.  I think you misunderstood the OP's question.

I'm assuming you understand that as sensor resolution increases, the 1:1 view's magnification also increases, right?   So looking at 1:1 with a D3200, you're looking at an area of your photo 1/4 the size of a 1:1 view with the D90.

So, I have a hunch that a 24MP APS-C sensor may have pixels smaller than the 17-55's smallest circle of confusion.  Maybe the 16MP does, too.  Meaning that if you zoom 1:1 with 24 megapixels, it might not look as "sharp" as viewing 1:1 with 12 megapixels.

No. The quality of the lens has nothing to do with either the size of the circle of confusion or the size of the airy disc. Both are functions of the aperture and other parameters, such as focal length, viewing size, etc. You're on the right trail, sort of, in that the 12mp sensor's 1:1 (or 100%) view might look better than the 24mp's view, but that can be misleading because to get to a 100% view, the 12mp sensor's image isn't magnified as much, but if the magnifications of images produced by both sensors are equalized so that the subjects in the frame are the same size, the 24mp image will be better, irregardless of the lens used, as long as the same lens is used on both cameras.

No.  Again, I think you misunderstood the OP's question.  The quality of the lens has everything to do with the size of the circle of least confusion.  A lens system whose projection produces high spherical aberrations (at a given aperture and focal length) for example, will produce a larger circle of least confusion than a lens with highly corrected spherical aberrations.

MarkJH was 100% correct that, if that circle of least confusion happens to be larger than the pixel pitch on which it's projected, a 1:1 view will not appear crisp.  But he was also 100% correct to suggest that this doesn't really matter as a measure or critique of sharpness for most printing and viewing situations--a point your read-and-response neglected to address.

Basically, you didn't read his (rather elegant) explanation very carefully, jumped on his use of terms (incorrectly) and blew your own wide "circle of confusion" over one of the more cogent answers to this sort of question I've ever read.

But then you commit a serious technical foul:

Here's a quote from Thom Hogan's S5 Pro review where among other things he compares its resolution with Nikon's 10mp D200. Following that is a link to his "How Big Can You Print?" article.

That's right, you quote freakin' Thom Hogan as some kind of authority on . . . something.  Thom is the Ron Burgundy of the photographic blogosphere: his posts "smell of finely bound leather books and rich mahogany," but seldom include a worthwhile photograph and never link to a professional portfolio of any kind.  "I'm kind of a big deal," his language says, but his photography--or complete lack thereof--definitely says otherwise.

Man, anyone can have opinions about gear, but Thom's results (like this week's Muskrat Love blog header) suggest his aren't exactly of the "authoritative" variety.


 MiraShootsNikon's gear list:MiraShootsNikon's gear list
Nikon D750 Nikon D850 Sony a7R III Nikon AF Nikkor 135mm f/2D DC Carl Zeiss Distagon T* 2,8/21 +12 more
Post (hide subjects) Posted by
(unknown member)
(unknown member)
(unknown member)
(unknown member)
(unknown member)
(unknown member)
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow