Lusting for fullframe (cont'd)

[Gary from Seattle]

No. Actually FF and m43rds are mostly the same if you take the same shot. It is not mostly better unless you can use a lower ISO or a larger aperture with more shallow DoF. If neither of those options are available what you said is wrong.

It is like a religion this belief.

Is the matching of DoF such an important consideration?

It can be. It isn't, for me, with landscapes, but it is, for me, with most of what I do, which is close-up/macro of invertebrates and botanical subjects.

I understand, and I am certainly not denying that concern for wide DoF is paramount to many photographic situations.

I shoot FF, APS-C and MFT with subjects ranging from deep space nebulae to aphids, and from fairywrens to elephants, and I have never been in a situation where I have chosen MFT or wished that I had used MFT over FF because it offered wider DoF for a given F number.

peter

Similar for me. I shoot 1/2.3", m43 and APS-C. I can get (and use, routinely for invertebrates and occasionally for botanical subjects) the same maximum DoF of f/45 equivalent with all of them. I now have FF too, and that would not provide as much DoF, but having now done comparison testing I don't envisage using it for invertebrates or botanical subjects (or common birds in flight, another of my areas of interest), only, possibly, subject to further comparison testing when some decent clouds and light return, for estuarine water/cloudscapes/sunsets, for which DoF isn't an issue.

Boy, I don't agree on landscapes at all. DOF is crucial in many of the landscape images I shoot in mountains, deserts, and rain forests. I just briefly glanced back at three day trips I've done in the past ten days or so and in 19 of 73 images I took DOF was an important consideration and in the best 3-5 images was hypercritical. In fact I discarded a fair number of these images just for the reason that I did not obtain critical focus on certain landscape images. In my gallery a quick glance also revealed that about 22 of some 60+ images had significant DOF challenges, many of which were maxed out without going to an absurd aperture with major diffraction.

With your camera between f/4 to f/5.6 is where it is at,

That is poor English and so it is hard to know what you are commenting about.

https://photographylife.com/best-mirrorless-cameras-you-can-buy-ranked

Telephoto images are, of course, a different case and many are essentially snap shots of some scene in the distance unless compression is used artistically. With regard to the latter, the same gallery contains about 15 images shot with various telephotos lengths where it was essentially flat field photography and DOF was unimportant. These images could have been and probably were shot at F4 to F5.6.

 

[David]

In the past, we used to determine how much DOF we can use "based" on what kind of output we are planning to display on and that is usually a print. And we usually shoot beyond diffraction because big prints at the time was of lower resolution. This means, while diffraction tends to rob away lens resolution, the loss of resolution is not high enough to affect print quality, due to the limited print resolution at that time. This was what many greats like Galen Rowell and Moose Petersen used to taught us in the past, so shooting like f/20 and higher. Jay Dickman, Olympus Visionary routinely shoot past MFT diffraction of f/8 to f/16 and f/20. So using full frame and achieving wide DOF is NOT a problem in the past and it is also not in the present.

What has changed is how we now view images. We no longer view them on prints. We view them on our 4k or 5k monitors mainly at 100%. That's like blowing it up past 40x60 prints. And in that case though, you would want to shoot with a lower f/stop say f/4 and f/5.6 to maintain the lens' high resolution on the center and corners. And then you would use a technique, if you're using a wide and medium zoom called focus stacking. The Nikon D850 has a focus stacking feature for both landscape and macro. The OMD and later Olympus cameras also has focus stacking. This way, you can have an infinite amount of DOF shot at the lens' highest resolving power; usually about 1 to 2 stop from maximum aperture. I do this some of the time and I use a software called Helicon Focus to stack them all together. The result is simply amazing and it can be printed big!

The second method is called stitching with a telephoto lens. This is desirable if you want to create a panoramic or wide angle view without the wide angle distortion as telephoto lenses create a flat field narrow view. Again, shot in pieces and then stitch them together in Photoshop can yield a much higher resolution image with excellent lens resolving power. You do need a nice tripod head that helps with smooth and accurate panning.

Since you will be using a tripod and a nice stiff head doing this, then the advantage of full frame shines, because at lower ISOs, full frame can achieve up to 15 stops of dynamic range and has much higher tonal range (smoother color transition) and of course lower noise at lower ISO.

 

[howardfuhrman]

Late after yesterdayI went to a photo exhibit in Los Angeles of Joe Santore who is shoots for National Geographic and is documenting all the animal species on earth. His photos are outstanding. I noted that that nearly all of the animal backgrounds were either black or white. There was a video narrated by Joe showing how he shot his photos. He has been visiting zoos around the world. He brings his own back drops and high power multiple flash lighting allowing in to shoot at small lens apertures for a wider depth of field. In his video, it showed he was shooting with a FF Nikon 810. I am not sure that the quality of his large prints would have been as good shooting with a smaller sensor camera.

Bear in mind Joe is a pro and based on the prints displayed at the exhibit, he may be one of the best pro's shooting animals, if not the best. Weight of his camera and lens does not apply to him as he is hauling back drops, large lighting and multiple lenses.

Again if the OP wants a FF camera, I would encourage the OP to get one.

 

[CanonSharpShooter]

Marketing will prevail in the end. These manufacturers will push ff cameras so low that it will be more expensive to shoot with m43 in the future. Than peoples preferences will change and don’t mind the size and weight of ff.

 

[bobn2]

In particular Bobn2 in response to a post by Jim Sterling noted a benefit to applying noise reduction to the higher resolution fullframe image before downresolving it to the mFT image size. I responded that it's, of course, true that applying extra noise reduction at any point in the post/processing chain will yield a more noise-free image BUT at the expense of some loss of resolution as well.

The principle is quite simple. A low pass filter will reduce noise. Since downsampling, properly done, includes a low pass filter, downsampling reduces noise (if you do it without the low pass filter, you replace the noise by aliasing). Noise reduction is supposed to reduce noise more than an equivalent amount of low-pass filtering. If it didn't, there wouldn't be much point having sophisticated NR, you'd just low pass. So, if instead of low-pass filtering you use NR to a level that reduces the detail to that equivalent to the low-pass filter, then you should have a lower proportion of noise. Simple as that, really.

 

[gx85sacto]

You ''downresolve" the noise, you downresolve the signal as well; you don't gain on snr by blowing up or shrinking down. Once you compensated the ff advantage by bringing iso up by 2 stops, the two images will look about the same no matter what you do.

 

[bobn2]

You ''downresolve" the noise, you downresolve the signal as well; you don't gain on snr by blowing up or shrinking down.

Actually you do by 'shrinking down', if you do the downsample properly - that is, with a low pass filter. 'Noise' and 'Signal' aren't separate things. 'Noise' is just a random variation of the 'signal' around an expected value. In mathematical terms, signal is autocorrelated, whereas noise, being random, isn't. That means that noise tends to average out on downsampling whilst signal doesn't.

 

[Sergey_Green]

No. Actually FF and m43rds are mostly the same if you take the same shot. It is not mostly better unless you can use a lower ISO or a larger aperture with more shallow DoF. If neither of those options are available what you said is wrong.

It is like a religion this belief.

Is the matching of DoF such an important consideration?

It can be. It isn't, for me, with landscapes, but it is, for me, with most of what I do, which is close-up/macro of invertebrates and botanical subjects.

I understand, and I am certainly not denying that concern for wide DoF is paramount to many photographic situations.

I shoot FF, APS-C and MFT with subjects ranging from deep space nebulae to aphids, and from fairywrens to elephants, and I have never been in a situation where I have chosen MFT or wished that I had used MFT over FF because it offered wider DoF for a given F number.

peter

Similar for me. I shoot 1/2.3", m43 and APS-C. I can get (and use, routinely for invertebrates and occasionally for botanical subjects) the same maximum DoF of f/45 equivalent with all of them. I now have FF too, and that would not provide as much DoF, but having now done comparison testing I don't envisage using it for invertebrates or botanical subjects (or common birds in flight, another of my areas of interest), only, possibly, subject to further comparison testing when some decent clouds and light return, for estuarine water/cloudscapes/sunsets, for which DoF isn't an issue.

Boy, I don't agree on landscapes at all. DOF is crucial in many of the landscape images I shoot in mountains, deserts, and rain forests. I just briefly glanced back at three day trips I've done in the past ten days or so and in 19 of 73 images I took DOF was an important consideration and in the best 3-5 images was hypercritical. In fact I discarded a fair number of these images just for the reason that I did not obtain critical focus on certain landscape images. In my gallery a quick glance also revealed that about 22 of some 60+ images had significant DOF challenges, many of which were maxed out without going to an absurd aperture with major diffraction.

With your camera between f/4 to f/5.6 is where it is at,

That is poor English and so it is hard to know what you are commenting about.

I thought you were a photographer, link is below,

https://photographylife.com/best-mirrorless-cameras-you-can-buy-ranked

Telephoto images are, of course, a different case and many are essentially snap shots of some scene in the distance unless compression is used artistically. With regard to the latter, the same gallery contains about 15 images shot with various telephotos lengths where it was essentially flat field photography and DOF was unimportant. These images could have been and probably were shot at F4 to F5.6.

 

[gardenersassistant]

No. Actually FF and m43rds are mostly the same if you take the same shot. It is not mostly better unless you can use a lower ISO or a larger aperture with more shallow DoF. If neither of those options are available what you said is wrong.

It is like a religion this belief.

Is the matching of DoF such an important consideration?

It can be. It isn't, for me, with landscapes, but it is, for me, with most of what I do, which is close-up/macro of invertebrates and botanical subjects.

I understand, and I am certainly not denying that concern for wide DoF is paramount to many photographic situations.

I shoot FF, APS-C and MFT with subjects ranging from deep space nebulae to aphids, and from fairywrens to elephants, and I have never been in a situation where I have chosen MFT or wished that I had used MFT over FF because it offered wider DoF for a given F number.

peter

Similar for me. I shoot 1/2.3", m43 and APS-C. I can get (and use, routinely for invertebrates and occasionally for botanical subjects) the same maximum DoF of f/45 equivalent with all of them. I now have FF too, and that would not provide as much DoF, but having now done comparison testing I don't envisage using it for invertebrates or botanical subjects (or common birds in flight, another of my areas of interest), only, possibly, subject to further comparison testing when some decent clouds and light return, for estuarine water/cloudscapes/sunsets, for which DoF isn't an issue.

Boy, I don't agree on landscapes at all. DOF is crucial in many of the landscape images I shoot in mountains, deserts, and rain forests. I just briefly glanced back at three day trips I've done in the past ten days or so and in 19 of 73 images I took DOF was an important consideration and in the best 3-5 images was hypercritical. In fact I discarded a fair number of these images just for the reason that I did not obtain critical focus on certain landscape images. In my gallery a quick glance also revealed that about 22 of some 60+ images had significant DOF challenges, many of which were maxed out without going to an absurd aperture with major diffraction.

With your camera between f/4 to f/5.6 is where it is at,

That is poor English and so it is hard to know what you are commenting about.

I thought you were a photographer, link is below,

I read the linked page and found an article titled "Best Mirrorless Cameras You Can Buy in 2018, Ranked". There were 15 references to f-numbers in the article, each associated with an image. The number of times each referenced f-number was used was:

f/1.2, 1
f/2.8, 2
f/4, 1
f/5.6, 5
f/8, 2
f/11, 2
f/13, 1
f/14, 1

So of those 15 photos, 6 used f/4 to f/5.6 and 9 used other apertures. So in terms of those photos I couldn't understand your "f/4 to f/5.6 is where it is at".

Your "for your camera" puzzled me too. The only camera mentioned in the signature of Gary from Seattle, to whom you were responding, is an Olympus E-M1. But the article you linked to had nine cameras which were medium format, full frame or APS-C, and only one micro four thirds camera, the Panasonic G9. The entry for the Panasonic G9 was the only one without an associated image, and so there was no mention of f-numbers in relation to any micro four thirds camera. So what relationship these images bear to Gary from Seattle's camera is mysterious to me.

Is it perhaps that you were referring only to the landscape images in that article? There were 10 of these, of which 5 used f/4 or f/5.6 and the other 5 used f/8, f/11 and f/14. The fact that half did and half did not use f/4 to f/5.6 does not seem to me to demonstrate that f/4 to f/5.6 is where it is at for the landscape images in that article. And in any case, I am puzzled why the images in that particular article should be taken to be definitive as to where it is at, for landscape or generally. And of course if you had been referring to the landscape images in that article that would still not explain the "for your camera" reference.

I obviously must have missed something here which provides the context that gives meaning to your comment "With your camera between f/4 to f/5.6 is where it is at". I would be very interested to learn what it is.

https://photographylife.com/best-mirrorless-cameras-you-can-buy-ranked

Telephoto images are, of course, a different case and many are essentially snap shots of some scene in the distance unless compression is used artistically. With regard to the latter, the same gallery contains about 15 images shot with various telephotos lengths where it was essentially flat field photography and DOF was unimportant. These images could have been and probably were shot at F4 to F5.6.

 

[knickerhawk]

In particular Bobn2 in response to a post by Jim Sterling noted a benefit to applying noise reduction to the higher resolution fullframe image before downresolving it to the mFT image size. I responded that it's, of course, true that applying extra noise reduction at any point in the post/processing chain will yield a more noise-free image BUT at the expense of some loss of resolution as well.

The principle is quite simple. A low pass filter will reduce noise. Since downsampling, properly done, includes a low pass filter, downsampling reduces noise (if you do it without the low pass filter, you replace the noise by aliasing). Noise reduction is supposed to reduce noise more than an equivalent amount of low-pass filtering. If it didn't, there wouldn't be much point having sophisticated NR, you'd just low pass. So, if instead of low-pass filtering you use NR to a level that reduces the detail to that equivalent to the low-pass filter, then you should have a lower proportion of noise. Simple as that, really.

Unfortunately, it's not that simple, really. You haven't factored in the impact of combining the NR with the chosen downsampling algorithm, which as you note will also include a low pass filter. The question I'm interested in is a practical one based on real tools such as ACR and Photoshop and how much of a visible IQ difference (if any) it makes when your workflow is NR > downsample vs. downsample > NR. I've now done quite a bit of comparisons over the past couple of days and, at least with two common NR options in PS/ACR and using various downsampling options, I can't say that it's always more effective to NR first before downsampling. Below is my attempt to clearly illustrate the principles at play. The left-most cross is from the D850 ISO6400 shot captured before any downsampling or NR has been applied. The second to left cross has been bicubic downsampled only (no NR applied). The visibly decreased noise with minimal aliasing indicates that the bicubic algorithm includes a low pass filter as you've described. The third and fourth crosses have both been bicubic downsampled. One of them had NR applied before downsampling and one of them had the same amount of NR applied after. Would you say that they show the same S/N or would you say that one is more successful?

 

[Adielle]

The result on the right is, of course, better. You lost too much of the original image on the left one, and added massive blur. The left result must be the one with noise reduction applied after resampling, unless you're doing or the algorithm is doing something wrong. Denoising before applying additional interpolation is the only way to get an optimal result.

 

[Sergey_Green]

No. Actually FF and m43rds are mostly the same if you take the same shot. It is not mostly better unless you can use a lower ISO or a larger aperture with more shallow DoF. If neither of those options are available what you said is wrong.

It is like a religion this belief.

Is the matching of DoF such an important consideration?

It can be. It isn't, for me, with landscapes, but it is, for me, with most of what I do, which is close-up/macro of invertebrates and botanical subjects.

I understand, and I am certainly not denying that concern for wide DoF is paramount to many photographic situations.

I shoot FF, APS-C and MFT with subjects ranging from deep space nebulae to aphids, and from fairywrens to elephants, and I have never been in a situation where I have chosen MFT or wished that I had used MFT over FF because it offered wider DoF for a given F number.

peter

Similar for me. I shoot 1/2.3", m43 and APS-C. I can get (and use, routinely for invertebrates and occasionally for botanical subjects) the same maximum DoF of f/45 equivalent with all of them. I now have FF too, and that would not provide as much DoF, but having now done comparison testing I don't envisage using it for invertebrates or botanical subjects (or common birds in flight, another of my areas of interest), only, possibly, subject to further comparison testing when some decent clouds and light return, for estuarine water/cloudscapes/sunsets, for which DoF isn't an issue.

Boy, I don't agree on landscapes at all. DOF is crucial in many of the landscape images I shoot in mountains, deserts, and rain forests. I just briefly glanced back at three day trips I've done in the past ten days or so and in 19 of 73 images I took DOF was an important consideration and in the best 3-5 images was hypercritical. In fact I discarded a fair number of these images just for the reason that I did not obtain critical focus on certain landscape images. In my gallery a quick glance also revealed that about 22 of some 60+ images had significant DOF challenges, many of which were maxed out without going to an absurd aperture with major diffraction.

With your camera between f/4 to f/5.6 is where it is at,

That is poor English and so it is hard to know what you are commenting about.

I thought you were a photographer, link is below,

I read the linked page and found an article titled "Best Mirrorless Cameras You Can Buy in 2018, Ranked". There were 15 references to f-numbers in the article, each associated with an image. The number of times each referenced f-number was used was:

f/1.2, 1
f/2.8, 2
f/4, 1
f/5.6, 5
f/8, 2
f/11, 2
f/13, 1
f/14, 1

So of those 15 photos, 6 used f/4 to f/5.6 and 9 used other apertures. So in terms of those photos I couldn't understand your "f/4 to f/5.6 is where it is at".

Your "for your camera" puzzled me too. The only camera mentioned in the signature of Gary from Seattle, to whom you were responding, is an Olympus E-M1. But the article you linked to had nine cameras which were medium format, full frame or APS-C, and only one micro four thirds camera, the Panasonic G9. The entry for the Panasonic G9 was the only one without an associated image, and so there was no mention of f-numbers in relation to any micro four thirds camera. So what relationship these images bear to Gary from Seattle's camera is mysterious to me.

Is it perhaps that you were referring only to the landscape images in that article? There were 10 of these, of which 5 used f/4 or f/5.6 and the other 5 used f/8, f/11 and f/14. The fact that half did and half did not use f/4 to f/5.6 does not seem to me to demonstrate that f/4 to f/5.6 is where it is at for the landscape images in that article. And in any case, I am puzzled why the images in that particular article should be taken to be definitive as to where it is at, for landscape or generally. And of course if you had been referring to the landscape images in that article that would still not explain the "for your camera" reference.

I obviously must have missed something here which provides the context that gives meaning to your comment "With your camera between f/4 to f/5.6 is where it is at". I would be very interested to learn what it is.

F/11 on the full frame camera is equivalent to f/5.6 on the four-thirds format, which is what Gary is using. There is one landscape frame with the medium format Fujifilm GFX 50S @f/11. The APS-C landscape frames are @f/5.6. All in all it averages to around f/4-f/5.6 on the mFT. Depending on the focal length Gary is using of course, I assume those would be more or less equivalent as well.

No, those are not the examples that everyone should follow, but they give a good hint of where the best settings for each format are. It is what the EXIF is for, is not it?

https://photographylife.com/best-mirrorless-cameras-you-can-buy-ranked

Telephoto images are, of course, a different case and many are essentially snap shots of some scene in the distance unless compression is used artistically. With regard to the latter, the same gallery contains about 15 images shot with various telephotos lengths where it was essentially flat field photography and DOF was unimportant. These images could have been and probably were shot at F4 to F5.6.

--
- sergey

 

[Great Bustard]

In particular Bobn2 in response to a post by Jim Sterling noted a benefit to applying noise reduction to the higher resolution fullframe image before downresolving it to the mFT image size. I responded that it's, of course, true that applying extra noise reduction at any point in the post/processing chain will yield a more noise-free image BUT at the expense of some loss of resolution as well.

The principle is quite simple. A low pass filter will reduce noise. Since downsampling, properly done, includes a low pass filter, downsampling reduces noise (if you do it without the low pass filter, you replace the noise by aliasing). Noise reduction is supposed to reduce noise more than an equivalent amount of low-pass filtering. If it didn't, there wouldn't be much point having sophisticated NR, you'd just low pass. So, if instead of low-pass filtering you use NR to a level that reduces the detail to that equivalent to the low-pass filter, then you should have a lower proportion of noise. Simple as that, really.

Unfortunately, it's not that simple, really. You haven't factored in the impact of combining the NR with the chosen downsampling algorithm, which as you note will also include a low pass filter. The question I'm interested in is a practical one based on real tools such as ACR and Photoshop and how much of a visible IQ difference (if any) it makes when your workflow is NR > downsample vs. downsample > NR.

Always good to put theories to the test with examples.

I've now done quite a bit of comparisons over the past couple of days and, at least with two common NR options in PS/ACR and using various downsampling options, I can't say that it's always more effective to NR first before downsampling. Below is my attempt to clearly illustrate the principles at play. The left-most cross is from the D850 ISO6400 shot captured before any downsampling or NR has been applied. The second to left cross has been bicubic downsampled only (no NR applied). The visibly decreased noise with minimal aliasing indicates that the bicubic algorithm includes a low pass filter as you've described. The third and fourth crosses have both been bicubic downsampled. One of them had NR applied before downsampling and one of them had the same amount of NR applied after. Would you say that they show the same S/N or would you say that one is more successful?

To my eyes, the SNR of the last two are all but identical. However, the last is significantly sharper with regards to the vertical bar of the +. That said, depending on the actual widths of the bars, the second to last my be the more accurate representation.

 

[bobn2]

In particular Bobn2 in response to a post by Jim Sterling noted a benefit to applying noise reduction to the higher resolution fullframe image before downresolving it to the mFT image size. I responded that it's, of course, true that applying extra noise reduction at any point in the post/processing chain will yield a more noise-free image BUT at the expense of some loss of resolution as well.

The principle is quite simple. A low pass filter will reduce noise. Since downsampling, properly done, includes a low pass filter, downsampling reduces noise (if you do it without the low pass filter, you replace the noise by aliasing). Noise reduction is supposed to reduce noise more than an equivalent amount of low-pass filtering. If it didn't, there wouldn't be much point having sophisticated NR, you'd just low pass. So, if instead of low-pass filtering you use NR to a level that reduces the detail to that equivalent to the low-pass filter, then you should have a lower proportion of noise. Simple as that, really.

Unfortunately, it's not that simple, really. You haven't factored in the impact of combining the NR with the chosen downsampling algorithm, which as you note will also include a low pass filter. The question I'm interested in is a practical one based on real tools such as ACR and Photoshop and how much of a visible IQ difference (if any) it makes when your workflow is NR > downsample vs. downsample > NR. I've now done quite a bit of comparisons over the past couple of days and, at least with two common NR options in PS/ACR and using various downsampling options, I can't say that it's always more effective to NR first before downsampling. Below is my attempt to clearly illustrate the principles at play. The left-most cross is from the D850 ISO6400 shot captured before any downsampling or NR has been applied. The second to left cross has been bicubic downsampled only (no NR applied). The visibly decreased noise with minimal aliasing indicates that the bicubic algorithm includes a low pass filter as you've described. The third and fourth crosses have both been bicubic downsampled. One of them had NR applied before downsampling and one of them had the same amount of NR applied after. Would you say that they show the same S/N or would you say that one is more successful?

To my eyes the one on the right has both a better SNR and also a greater high frequency component. I also wouldn't be at all surprised if you pop up and tell us that's the one which has been NRed after downsampling, because it looks visibly more like the simple downsample. However, designing experiments is complicated, and there are always caveats unless very carefully done, so even if that is the case, with as many uncontrolled variables as there are, and with a single and simple test case, I wouldn't say that this necessarily invalidated the simple theory above. As for your speculation about 'you haven't factored in the impact of combining the NR with the chosen downsampling algorithm' it's a good point, but one would need to be more specific about what that impact might be, and how the two might interact negatively.

--
Ride easy, William.
Bob

 

[gardenersassistant]

No. Actually FF and m43rds are mostly the same if you take the same shot. It is not mostly better unless you can use a lower ISO or a larger aperture with more shallow DoF. If neither of those options are available what you said is wrong.

It is like a religion this belief.

Is the matching of DoF such an important consideration?

It can be. It isn't, for me, with landscapes, but it is, for me, with most of what I do, which is close-up/macro of invertebrates and botanical subjects.

I understand, and I am certainly not denying that concern for wide DoF is paramount to many photographic situations.

I shoot FF, APS-C and MFT with subjects ranging from deep space nebulae to aphids, and from fairywrens to elephants, and I have never been in a situation where I have chosen MFT or wished that I had used MFT over FF because it offered wider DoF for a given F number.

peter

Similar for me. I shoot 1/2.3", m43 and APS-C. I can get (and use, routinely for invertebrates and occasionally for botanical subjects) the same maximum DoF of f/45 equivalent with all of them. I now have FF too, and that would not provide as much DoF, but having now done comparison testing I don't envisage using it for invertebrates or botanical subjects (or common birds in flight, another of my areas of interest), only, possibly, subject to further comparison testing when some decent clouds and light return, for estuarine water/cloudscapes/sunsets, for which DoF isn't an issue.

Boy, I don't agree on landscapes at all. DOF is crucial in many of the landscape images I shoot in mountains, deserts, and rain forests. I just briefly glanced back at three day trips I've done in the past ten days or so and in 19 of 73 images I took DOF was an important consideration and in the best 3-5 images was hypercritical. In fact I discarded a fair number of these images just for the reason that I did not obtain critical focus on certain landscape images. In my gallery a quick glance also revealed that about 22 of some 60+ images had significant DOF challenges, many of which were maxed out without going to an absurd aperture with major diffraction.

With your camera between f/4 to f/5.6 is where it is at,

That is poor English and so it is hard to know what you are commenting about.

I thought you were a photographer, link is below,

I read the linked page and found an article titled "Best Mirrorless Cameras You Can Buy in 2018, Ranked". There were 15 references to f-numbers in the article, each associated with an image. The number of times each referenced f-number was used was:

f/1.2, 1
f/2.8, 2
f/4, 1
f/5.6, 5
f/8, 2
f/11, 2
f/13, 1
f/14, 1

So of those 15 photos, 6 used f/4 to f/5.6 and 9 used other apertures. So in terms of those photos I couldn't understand your "f/4 to f/5.6 is where it is at".

Your "for your camera" puzzled me too. The only camera mentioned in the signature of Gary from Seattle, to whom you were responding, is an Olympus E-M1. But the article you linked to had nine cameras which were medium format, full frame or APS-C, and only one micro four thirds camera, the Panasonic G9. The entry for the Panasonic G9 was the only one without an associated image, and so there was no mention of f-numbers in relation to any micro four thirds camera. So what relationship these images bear to Gary from Seattle's camera is mysterious to me.

Is it perhaps that you were referring only to the landscape images in that article? There were 10 of these, of which 5 used f/4 or f/5.6 and the other 5 used f/8, f/11 and f/14. The fact that half did and half did not use f/4 to f/5.6 does not seem to me to demonstrate that f/4 to f/5.6 is where it is at for the landscape images in that article. And in any case, I am puzzled why the images in that particular article should be taken to be definitive as to where it is at, for landscape or generally. And of course if you had been referring to the landscape images in that article that would still not explain the "for your camera" reference.

I obviously must have missed something here which provides the context that gives meaning to your comment "With your camera between f/4 to f/5.6 is where it is at". I would be very interested to learn what it is.

F/11 on the full frame camera is equivalent to f/5.6 on the four-thirds format, which is what Gary is using. There is one landscape frame with the medium format Fujifilm GFX 50S @f/11. The APS-C landscape frames are @f/5.6. All in all it averages to around f/4-f/5.6 on the mFT.

Understood. I see that 7 out of the 10 are in the f/4 to f/5.6 m43 equivalent range. Thanks.

Depending on the focal length Gary is using of course, I assume those would be more or less equivalent as well.

No, those are not the examples that everyone should follow, but they give a good hint of where the best settings for each format are. It is what the EXIF is for, is not it?

https://photographylife.com/best-mirrorless-cameras-you-can-buy-ranked

Telephoto images are, of course, a different case and many are essentially snap shots of some scene in the distance unless compression is used artistically. With regard to the latter, the same gallery contains about 15 images shot with various telephotos lengths where it was essentially flat field photography and DOF was unimportant. These images could have been and probably were shot at F4 to F5.6.

 

[Jonas Palm]

In particular Bobn2 in response to a post by Jim Sterling noted a benefit to applying noise reduction to the higher resolution fullframe image before downresolving it to the mFT image size. I responded that it's, of course, true that applying extra noise reduction at any point in the post/processing chain will yield a more noise-free image BUT at the expense of some loss of resolution as well.

The principle is quite simple. A low pass filter will reduce noise. Since downsampling, properly done, includes a low pass filter, downsampling reduces noise (if you do it without the low pass filter, you replace the noise by aliasing). Noise reduction is supposed to reduce noise more than an equivalent amount of low-pass filtering. If it didn't, there wouldn't be much point having sophisticated NR, you'd just low pass. So, if instead of low-pass filtering you use NR to a level that reduces the detail to that equivalent to the low-pass filter, then you should have a lower proportion of noise. Simple as that, really.

Unfortunately, it's not that simple, really. You haven't factored in the impact of combining the NR with the chosen downsampling algorithm, which as you note will also include a low pass filter. The question I'm interested in is a practical one based on real tools such as ACR and Photoshop and how much of a visible IQ difference (if any) it makes when your workflow is NR > downsample vs. downsample > NR. I've now done quite a bit of comparisons over the past couple of days and, at least with two common NR options in PS/ACR and using various downsampling options, I can't say that it's always more effective to NR first before downsampling. Below is my attempt to clearly illustrate the principles at play. The left-most cross is from the D850 ISO6400 shot captured before any downsampling or NR has been applied. The second to left cross has been bicubic downsampled only (no NR applied). The visibly decreased noise with minimal aliasing indicates that the bicubic algorithm includes a low pass filter as you've described. The third and fourth crosses have both been bicubic downsampled. One of them had NR applied before downsampling and one of them had the same amount of NR applied after. Would you say that they show the same S/N or would you say that one is more successful?

To my eyes the one on the right has both a better SNR and also a greater high frequency component. I also wouldn't be at all surprised if you pop up and tell us that's the one which has been NRed after downsampling, because it looks visibly more like the simple downsample. However, designing experiments is complicated, and there are always caveats unless very carefully done, so even if that is the case, with as many uncontrolled variables as there are, and with a single and simple test case, I wouldn't say that this necessarily invalidated the simple theory above. As for your speculation about 'you haven't factored in the impact of combining the NR with the chosen downsampling algorithm' it's a good point, but one would need to be more specific about what that impact might be, and how the two might interact negatively.

--
Ride easy, William.
Bob

Assuming good downsampling, we are basically comparing different noise reduction methods. I’m postulating that you can generally do a better job on the original data, and then downsample, than a (good) automated process, and much better than a simplistic downsample.

However, different processes will lead to different artifacts, and the differences can’t be anything but quite subtle in a case like the selected example. There’s just not a whole lot of difference between the two originals.

Great Bustard had the right of it when he questioned why you would downsample the higher resolution original at all. For comparative purposes of course, but you do that by tossing away the resolution advantage of the high-res original, and potentially introducing artifacts in the process.

We’ll see a lot of automated downsampling with the upcoming 48MP cell phones. There might even (hopefully) be some meaty technical articles on the minutiae of the process.

 

[petreluk]

Just buy an FF camera and have done with it, if that floats your boat. It will do things an M43 camera cannot do. No amount of bizarre defensive techtalk will ever change that. I would not be in the least bit surprised if FF was the dominant format in five years’ time, with a couple of APS-C and 1” specials for the long-reach crowd and everyone else using very capable smartphones. I like M43 but Olympus versus rest of the world is a bit of a foregone conclusion, isn’t it.

--
==================
https://www.flickr.com/photos/petreluk/

 

[Gary from Seattle]

No. Actually FF and m43rds are mostly the same if you take the same shot. It is not mostly better unless you can use a lower ISO or a larger aperture with more shallow DoF. If neither of those options are available what you said is wrong.

It is like a religion this belief.

Is the matching of DoF such an important consideration?

It can be. It isn't, for me, with landscapes, but it is, for me, with most of what I do, which is close-up/macro of invertebrates and botanical subjects.

I understand, and I am certainly not denying that concern for wide DoF is paramount to many photographic situations.

I shoot FF, APS-C and MFT with subjects ranging from deep space nebulae to aphids, and from fairywrens to elephants, and I have never been in a situation where I have chosen MFT or wished that I had used MFT over FF because it offered wider DoF for a given F number.

peter

Similar for me. I shoot 1/2.3", m43 and APS-C. I can get (and use, routinely for invertebrates and occasionally for botanical subjects) the same maximum DoF of f/45 equivalent with all of them. I now have FF too, and that would not provide as much DoF, but having now done comparison testing I don't envisage using it for invertebrates or botanical subjects (or common birds in flight, another of my areas of interest), only, possibly, subject to further comparison testing when some decent clouds and light return, for estuarine water/cloudscapes/sunsets, for which DoF isn't an issue.

Boy, I don't agree on landscapes at all. DOF is crucial in many of the landscape images I shoot in mountains, deserts, and rain forests. I just briefly glanced back at three day trips I've done in the past ten days or so and in 19 of 73 images I took DOF was an important consideration and in the best 3-5 images was hypercritical. In fact I discarded a fair number of these images just for the reason that I did not obtain critical focus on certain landscape images. In my gallery a quick glance also revealed that about 22 of some 60+ images had significant DOF challenges, many of which were maxed out without going to an absurd aperture with major diffraction.

With your camera between f/4 to f/5.6 is where it is at,

That is poor English and so it is hard to know what you are commenting about.

I thought you were a photographer, link is below,

I read the linked page and found an article titled "Best Mirrorless Cameras You Can Buy in 2018, Ranked". There were 15 references to f-numbers in the article, each associated with an image. The number of times each referenced f-number was used was:

f/1.2, 1
f/2.8, 2
f/4, 1
f/5.6, 5
f/8, 2
f/11, 2
f/13, 1
f/14, 1

So of those 15 photos, 6 used f/4 to f/5.6 and 9 used other apertures. So in terms of those photos I couldn't understand your "f/4 to f/5.6 is where it is at".

Your "for your camera" puzzled me too. The only camera mentioned in the signature of Gary from Seattle, to whom you were responding, is an Olympus E-M1. But the article you linked to had nine cameras which were medium format, full frame or APS-C, and only one micro four thirds camera, the Panasonic G9. The entry for the Panasonic G9 was the only one without an associated image, and so there was no mention of f-numbers in relation to any micro four thirds camera. So what relationship these images bear to Gary from Seattle's camera is mysterious to me.

Is it perhaps that you were referring only to the landscape images in that article? There were 10 of these, of which 5 used f/4 or f/5.6 and the other 5 used f/8, f/11 and f/14. The fact that half did and half did not use f/4 to f/5.6 does not seem to me to demonstrate that f/4 to f/5.6 is where it is at for the landscape images in that article. And in any case, I am puzzled why the images in that particular article should be taken to be definitive as to where it is at, for landscape or generally. And of course if you had been referring to the landscape images in that article that would still not explain the "for your camera" reference.

I obviously must have missed something here which provides the context that gives meaning to your comment "With your camera between f/4 to f/5.6 is where it is at". I would be very interested to learn what it is.

F/11 on the full frame camera is equivalent to f/5.6 on the four-thirds format, which is what Gary is using. There is one landscape frame with the medium format Fujifilm GFX 50S @f/11. The APS-C landscape frames are @f/5.6. All in all it averages to around f/4-f/5.6 on the mFT. Depending on the focal length Gary is using of course, I assume those would be more or less equivalent as well.

There is a difference between where one would shoot for maximum IQ, and what aperture at some particular focal length one would use to assure focus in some particular scene with accordant DOF requirements. Yesterday for landscape I used F5.6 to F9 m4/3. Late in the day the light became really great for about one hour although I was in deep rain forest. Accordingly, and shooting on the run every 3-5 minutes, and concerned as at even ISO 1600 I was only able to get SS handheld at 1/5 to 1/25 for most shots, I made the mistake for awhile of using F6.3, and consequently on a number of shots failed on near focus. I violated my rule of shooting most landscape at F7.1 because I was in a hurry and most concerned about SS and choosing compositions; and I think initially there was one shot that I used F6.3 and thought I could squeak in DOF focus on that shot. Then, as I moved rapidly keeping my eyes open for compositions, forgot to reset to F7.1. Early in the day I shot two compositions at F5.6 and one failed. Along the way I shot one composition intentionally at F9, ISO 1600 FL 26mm as I had to basically shoot through a narrow window between trees and shrubs to cut out distractions. At the time, with subjects near, middle ground, and at a distance, I wanted to shoot what would look like an array of moss-covered trunks retreating into the distance with the last the real subject of the image. There was only one small place to stand to shoot this particular scene, and the closest tree I knew would not even be in focus at the F9 aperture I chose. I was also shooting while balancing precariously at 1/13 second so it is unlikely I could have used F11 (or that that aperture would have been sufficient for DOF focus anyway); and had I, I probably would not have been able to get a shot without some motion on my part. Even in more favorable circumstances on that shot I was limited by my positioning and necessary focal length for cropping.



Even in good light shooting summer wildflowers and landscape images, I often have to go to F8 to F11 for DOF even at short FL's because small wildflowers to be significant in the image need to be very near to make them large and hence power an image against a mountain or desert backdrop. More often than not I will get just as close as my positioning and DOF at a reasonable aperture will allow. I have any number of examples. With m4/3 comparing IQ, which is still near peak at F5.6, F8 loses appx 8% and F11 not more than 15% in resolution. Those losses pale in comparison to composition or reasonably sharp focus

No, those are not the examples that everyone should follow, but they give a good hint of where the best settings for each format are. It is what the EXIF is for, is not it?

https://photographylife.com/best-mirrorless-cameras-you-can-buy-ranked

Telephoto images are, of course, a different case and many are essentially snap shots of some scene in the distance unless compression is used artistically. With regard to the latter, the same gallery contains about 15 images shot with various telephotos lengths where it was essentially flat field photography and DOF was unimportant. These images could have been and probably were shot at F4 to F5.6.

 

[TN Args]

Just buy an FF camera and have done with it, if that floats your boat. It will do things an M43 camera cannot do. No amount of bizarre defensive techtalk will ever change that. I would not be in the least bit surprised if FF was the dominant format in five years’ time, with a couple of APS-C and 1” specials for the long-reach crowd and everyone else using very capable smartphones. I like M43 but Olympus versus rest of the world is a bit of a foregone conclusion, isn’t it.

In your mind it is. I presume you are a multi-billionaire, because you know how markets are going to move. Apparently it's all foregone conclusions. LOL

 

[bobn2]

In particular Bobn2 in response to a post by Jim Sterling noted a benefit to applying noise reduction to the higher resolution fullframe image before downresolving it to the mFT image size. I responded that it's, of course, true that applying extra noise reduction at any point in the post/processing chain will yield a more noise-free image BUT at the expense of some loss of resolution as well.

The principle is quite simple. A low pass filter will reduce noise. Since downsampling, properly done, includes a low pass filter, downsampling reduces noise (if you do it without the low pass filter, you replace the noise by aliasing). Noise reduction is supposed to reduce noise more than an equivalent amount of low-pass filtering. If it didn't, there wouldn't be much point having sophisticated NR, you'd just low pass. So, if instead of low-pass filtering you use NR to a level that reduces the detail to that equivalent to the low-pass filter, then you should have a lower proportion of noise. Simple as that, really.

Unfortunately, it's not that simple, really. You haven't factored in the impact of combining the NR with the chosen downsampling algorithm, which as you note will also include a low pass filter. The question I'm interested in is a practical one based on real tools such as ACR and Photoshop and how much of a visible IQ difference (if any) it makes when your workflow is NR > downsample vs. downsample > NR.

Always good to put theories to the test with examples.

I've now done quite a bit of comparisons over the past couple of days and, at least with two common NR options in PS/ACR and using various downsampling options, I can't say that it's always more effective to NR first before downsampling. Below is my attempt to clearly illustrate the principles at play. The left-most cross is from the D850 ISO6400 shot captured before any downsampling or NR has been applied. The second to left cross has been bicubic downsampled only (no NR applied). The visibly decreased noise with minimal aliasing indicates that the bicubic algorithm includes a low pass filter as you've described. The third and fourth crosses have both been bicubic downsampled. One of them had NR applied before downsampling and one of them had the same amount of NR applied after. Would you say that they show the same S/N or would you say that one is more successful?

To my eyes, the SNR of the last two are all but identical. However, the last is significantly sharper with regards to the vertical bar of the +. That said, depending on the actual widths of the bars, the second to last my be the more accurate representation.

Not sure about that. It appears to have an abrupt transition from 'light' to 'dark' on the right hand side of the vertical bar, which as John Sheehy will tel you, is something that should never happen in a properly sampled image. It looks like a downsampling artefact to me. Knickerhawk is using bicubic, which whilst it has a better inherent low-pass than nearest neighbour still doesn't have enough to prevent aliasing. Lanczos is a bit worse. The issue is really that these interpolation algorithms are a little bit anachronistic, they are designed to optimise upsampling and produce convincing interpolation, not to optimise downsampling.

--
Ride easy, William.
Bob