Resolution test, between crop (18MP) and fullframe (12.8MP)

Started Mar 12, 2013 | Discussions thread
Great Bustard Forum Pro • Posts: 41,786
Re: The test was performed properly.

ultimitsu wrote:

Great Bustard wrote:

ultimitsu wrote:

Great Bustard wrote:

Well, it was almost done properly. ...

You are looking at this test as how it emulates real world situation, I agree it can. But it was not set out to merely do so, it is set out to disapprove the theory that " with imperfect lens, at some point, 5D1 will out-resolve 60D." To that end, it is done completely wrong.

Ah. I was unaware that the purpose of the test was to prove that the 5Dc would outresolve the 60D at some point. Yes, it did fail in that purpose, as the test showed the opposite (though insignificantly so, in my opinion).

The purpose was not to prove "that the 5Dc would outresolve the 60D at some point". it was the opposite, the purpose was to disprove "that the 5Dc would outresolve the 60D at some point". it was to prove that hypothesis can never happen.

Ah.  Well, in that case, the OP simply provided an instance when the 60D outresolved the 5Dc (again, in my opinion, insignificantly so), but, of course, that does not prove it always will.

To prove this argument wrong, you need to shoot with identical imperfect lens. The test is wrong because 24-70II is quite close to a perfectly, made worse by it was less perfect on 5D than it is on 60D due to aperture difference.

All systems suffer the same from diffraction softening at the same DOF, but all lenses do not suffer from the same aberrations at the same DOF.

My view is that at F11 even without diffraction consideration, lens resolution power is reduced by virtue of so much information now passes through a much narrower section of glass.

There's no danger of photons colliding, even through very small apertures.

it is not about photons colliding. there is less diffraction at F2 than F4 yet most F1.4 lens is sharper at F4 than F2. it goes to show that resolution potential is not just linked with diffraction.

In terms of the lens, we have three things going on with respect to resolution as we stop down:

  1. Lessening lens aberrations.
  2. Greater DOF.
  3. Greater diffraction softening.

The lessening lens aberrations and greater DOF both serve to yield a more detailed photo, and the greater diffraction softening serves to soften the photo.

If we only consider the portions of the scene within the DOF, when the increasing diffraction softening outweighs the lessening lens aberrations will depend on the lens and where in the frame we are looking.  It is not uncommon for the corners to lag a stop behind the center in this regard.

If we bring DOF back into the equation, there is no set rule as to when having more of the scene within the DOF makes for an overall sharper photo than the increased diffraction softening.

Given how many landscape photographers are shooting at even f/16 on FF, it is ingenuous to cry "foul" for this particular test.

Just because people shoot F16 does not mean F16 is fine for the purpose for this test. People shoot F16 because for some scenes they must use F16 - the lens is far from perfect but the shot requires. What we are dealing here is whether 60D received unfair preferential treatment than 5D1 in what is purported to be a test.

The 60D did not "receive unfair preferential treatment". Perhaps you were looking for photos of a flat surface shot at the sharpest aperture of the lens at the optimum focal length from the optimum distance to prove something or other?

To prove what OP set out to prove, he needed to shoot a flat surface with a imperfect lens. for example 50 F1.8 at F1.8.

It would be odd to compare 50mm on FF to 50mm on crop.  I know Peter is fond of saying a photographer using FF and crop would both use a 50mm prime, and the FF photographer would just get closer, but I'm more of the opinion that the FF photographer would use 85mm.

This test is what one would call "rubbish in rubbish out".

Rather, it is a good example of a comparison of a deep scene at the same DOF.

It is a good example of a comparison of a deep scene at the same DOF, if comparison of a deep scene at the same DOF was the purpose. But it is not the purpose. For its purpose ( disapprove the theory that " with imperfect lens, at some point, 5D1 will out-resolve 60D."), it is "rubbish in rubbish out".

Let's just say that the photos are a decent demonstration of the resolution differences between a 5Dc and 60D for a scene with a relatively deep DOF.

You go on say that while I insist it is rubbish in rubbish out for what it set out to do. We are not in disagreement.


Post (hide subjects) Posted by
(unknown member)
MOD schmegg
MOD schmegg
MOD schmegg
MOD schmegg
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow