Pages: 1 [2]   Go Down

Author Topic: Thom Hogan (and thoughts of mine) on format equivalence...  (Read 3241 times)

Paul Roark

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 398
Re: Thom Hogan (and thoughts of mine) on format equivalence...
« Reply #20 on: May 04, 2019, 12:11:53 pm »

To me the issue is, in part, what is the smallest, lightest equipment that can allow me to meet my on-the-wall print sharpness standards.  For this real world goal, it appears to me that the impact of using software, and specifically Topaz's AI Gigapixel and/or Sharpen, is a major part of the equation.  Can APS-C now give results essentially equal, visually, to what "full frame 35mm" sensors previously produced?  (I use the Sony a7rii now.)

As a landscape shooter who likes everything very sharp, a significant and related issue is the DOF advantages of smaller formats.

An aps-c camera is in the mail to me now.  So, I'll be testing this soon to see how small and light I can travel/hike.

FWIW, in the film days, medium format was what I needed, and I used Tech Pan.  If a camera that fits into a pocket can equal my Rollei SL66, I'll be a happy camper/hiker.  (I am skeptical, but it's worth testing.)

Paul
www.PaulRoark.com
Logged

Aram Hăvărneanu

  • Full Member
  • ***
  • Offline Offline
  • Posts: 214
Re: Thom Hogan (and thoughts of mine) on format equivalence...
« Reply #21 on: May 04, 2019, 12:44:37 pm »

There is no DOF advantage of smaller formats. Unfortunately, this myth seems to persist because at first glance it is true. But it isn't. No format has ANY DOF advantage over any other format once you factor in diffraction.

For a given FOV, lenses with the same physical aperture (in mm, not f-stops) will have the same DOF and diffraction characteristics on any format, large and small.

Of course the same physical aperture translates to different f-stops on different formats (assuming lenses with the same FOV), so it appears that smaller formats have an advantage (because we think in f-stops, not in physical aperture sizes). But it's an illusion. Assuming you stop the lens down just to the point that diffraction is not an issue, you will find that you will have stepped it down to the same physical size on both small and large formats, and at that particular (same) physical size the DOF is the same on both formats.

Larger formats gather more total light though, giving better quality for a given ISO, but assuming you can stack shots and combine them later you can trade more exposure time for the same image quality.

Of course, larger format have more resolution, and potentially better lenses.
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: Thom Hogan (and thoughts of mine) on format equivalence...
« Reply #22 on: May 04, 2019, 02:54:04 pm »

There is no DOF advantage of smaller formats. Unfortunately, this myth seems to persist because at first glance it is true. But it isn't. No format has ANY DOF advantage over any other format once you factor in diffraction.

For a given FOV, lenses with the same physical aperture (in mm, not f-stops) will have the same DOF and diffraction characteristics on any format, large and small.
Agreed (mostly); this is a point I have tried to make in many discussions over the years: the equivalency discussed by Thom Hogan article also extends to diffraction and its affect on MTF and resolution, because at equal "effective aperture size" (entrance pupil diameter) the sizes of circles of confusion (OOF effects) and of Airy disks (diffraction effects) scale linearly with focal length and thus with the size of the image at the focal plane.  There are opposing myths that to some degree cancel out: more DOF with a smaller format vs less noise with a larger format; both often born from comparing at equal aperture ratio and equal ISO speed.

There is one slight advantage to a smaller format when stopping down for adequate DOF though, through reduced in-camera noise production (dark noise, read noise whatever): equivalent exposures are gathering the same total amount of light but due to the greater total area of photosites and their electron wells, a larger sensor adds more in-camera generated noise to the equal amounts of photon shot noise.

In this context, your comment that "Larger formats gather more total light" is slightly misleading; they can in some circumstances, but are not doing so with the equivalent choices of aperture, shutter speed and so on under discussion here. A larger format can only gather more light when it is used with some combination of larger aperture (more OOF blur) and longer exposure time (potentially more motion blur, from camera and/or subject).

And then as to "larger format have more resolution"; sometimes!
1) When resolution in one format is limited by the optimal trade-off between diffraction and OOF effects, that same resolution limit persists in any larger format, so then a larger format cannot offer significantly more resolution.

2) Achieving more resolution without sacrificing per-pixel IQ (SNR) also requires breaking equivalency in order to gather more light, and so more blur of the types mentioned above.

Maybe the strongest argument for a larger format is that it allows using a higher f-stop for "equivalent" composition, which can lead to lower lens aberration effects, with the extreme case being that formats below some size cannot have a lens of sufficiently low "equivalent" f-stop at all. Upper limits on effective aperture size is the most fundamental limit of what pocketable cameras like mobile phones are capable of.
Logged

Paul Roark

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 398
Re: Thom Hogan (and thoughts of mine) on format equivalence...
« Reply #23 on: May 04, 2019, 04:07:18 pm »

There is no DOF advantage of smaller formats. ...

See the demonstration at http://www.paulroark.com/DOF-50-v-75mm-f8.JPG

Paul
www.PaulRoark.com
Logged

Paul Roark

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 398
Re: Thom Hogan (and thoughts of mine) on format equivalence...
« Reply #24 on: May 04, 2019, 05:41:58 pm »

The following URL adds an image up-res'd in Topax to show the combination of better DOF with what Topax Gigapixel AI can do.

http://www.paulroark.com/DOF-50-v-75-v-50-up-res.jpg

The results are going to affect what I carry.  Software is affecting what is needed to reach a specific on-the-wall print sharpness target.

Paul
www.PaulRoark.com
« Last Edit: May 04, 2019, 07:50:26 pm by Paul Roark »
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: Thom Hogan (and thoughts of mine) on format equivalence...
« Reply #25 on: May 04, 2019, 10:16:51 pm »

There is no DOF advantage of smaller formats. Unfortunately, this myth seems to persist because at first glance it is true. But it isn't. No format has ANY DOF advantage over any other format once you factor in diffraction.

For a given FOV, lenses with the same physical aperture (in mm, not f-stops) will have the same DOF and diffraction characteristics on any format, large and small.


That's an interesting point, but it does need to be emphasized that there is no DOF advantage of smaller formats provided that everything else that is relevant remains the same, including the FOV in the image, the pixel pitch, the pixel quality of the sensor, and the lens quality and characteristics.

An obvious example would be the comparison between the 20 mp Nikon D500 (a cropped format) and the 45 mp D850 (full frame format). Both cameras appear to have the same quality of pixels (according to DXOMark), and the same density of pixels. The Nikon full frame sensor is 2.25x the area of their cropped DX format. 2.25 x 20 = 45.

Suppose we use both cameras with the same 50 mm lens, shooting the same scene at the same f/stop, using the same ISO and the same shutter speed. The image from the D850 will have a wider FOV, so in order to compare apples with apples it needs to be cropped to the same size as the D500 image. The cropped D850 image will then be identical to the D500 image, excluding the normal margins of error of course.

It will have the same DOF, same degree of sharpness, same amount of noise, same degree of overexposure or underexposure, and so on.
The smaller format in this example would appear to have no advantage at all, in terms of inherent image quality, although it clearly has an advantage in respect of weight, and possibly other software features which I haven't investigated.

However, the larger format of the D850, in this example, has the very obvious advantage of turning a standard 50mm lens into a 50-75mm zoom, from the perspective of the smaller DX format, and a zoom which is likely much better quality than the equivalent 33-50 mm zoom which would give the D500 the same range of FOVs as the D850 with 50 mm lens.
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: Thom Hogan (and thoughts of mine) on format equivalence...
« Reply #26 on: May 04, 2019, 10:45:19 pm »

See the demonstration at http://www.paulroark.com/DOF-50-v-75mm-f8.JPG

Paul
www.PaulRoark.com
No dispute that a smaller format/focal length has more DOF at equal aperture ratio, but a larger format can match both the DOF and shutter speed by increasing the ISO speed, and the result will give roughly the same noise level at equal resolution/image size.

This does however give up most of the image quality advantage that people are generally seeking when they accept the greater cost and bulk of a larger format. Perhaps it is better to say that the advantages in noise, dynamic range, resolution etc. usually associated with larger formats are achieved only when one accepts a trade-off of less DOF and/or longer exposure times.
Logged

faberryman

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4851
Re: Thom Hogan (and thoughts of mine) on format equivalence...
« Reply #27 on: May 05, 2019, 12:46:38 pm »

I am surprised that some posters say that they prefer the m43 format for its depth of field. Could they not stop down with a larger format - APS-C or FF? We know that beyond a certain point, they cannot open up for less depth of field, so their hands are tied.
Logged

Paul Roark

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 398
Re: Thom Hogan (and thoughts of mine) on format equivalence...
« Reply #28 on: May 05, 2019, 04:14:21 pm »

No dispute that a smaller format/focal length has more DOF at equal aperture ratio, ...

This does however give up most of the image quality advantage ... of a larger format. ...

My interest in aps-c sized cameras (and enlargement of that part of a full frame image) is to supplement and enhance my Sony a7rii full frame, and it is due to the success of Topaz in its AI Gigapixel software.  Possibly more effective DOF would just be a bonus.   

It looks to me that software (by Topaz) has reached the point where we can reasonably get what we thought of as "full frame" quality out of an aps-c size sensor (or enlargement of that size/part of a full frame image).  That is, I believe I can make a display print for the wall with an aps-c sensor that will be visually equal to what, before I tested Topaz'x AI Gigapixel, I would have assumed only "full frame 35" and above could achieve.

The first actual change is that I have stopped carrying around my 75 - 90 mm range lenses for my Sony.  That range is now better and more flexibly handled by my 50mm lens. 

The next (larger) question is whether a small, pocket aps-c that is "28mm equivalent" can replace my need to carry a wide angle lens for the Sony.  I think the real issue is going to be the quality of the lens and controls.  I'm skeptical, but I've ordered a small camera that some have said has an excellent lens.  I'll test it and send it back if it does not meet my criteria.

Paul
www.PaulRoark.com
Logged
Pages: 1 [2]   Go Up