Why I stop down

[Greg7579]

Jim doesn't care if I think it is more than 1 stop. Now [nor?] should you.

Jim does care. If you would establish quantitative metrics for how you judge DOF, we could probably agree on how many stops. But you have consistently refused to do that.

And you never did perform that test with the GF 45 that I challenged you to do.

If we have clear definitions, we could have more light and less heat. But in years of trying to get those from you, I've pretty much given up hope.

Don't give up hope JIm! I'm still here for you. That F9 shot of the kids.... That surprised me. I saw it in post and said to myself.... No way. That looks like FF DOF.

 

[Greg7579]

I would say that if you could (somehow) increase the per pixel sharpness of the in focus pixels, that would draw more attention to the out of focus pixels and result in less visible depth of field. This is equivalent to using a tighter CoC number in your calculator.

To get the calculated dof to agree with your subjective sense, you have to use an appropriate CoC value in your calculator, as Jim has patiently explained about to you about 10 thousand times. It doesn't mean there is something exceptional about MF sized sensors, it means you are using the wrong CoC value in your calculator. Use the right CoC value and, hey presto!, calculator and Mark I eyeballs agree.

The concept of DoF is based on acceptable level of unsharpness.

Yes Yes Yes!

If we except maximum sharpness, DoF will be essentially nil.

On a system with sharp lenses, the zone of maximum sharpness will be thinner that it would be with blurry lenses.

Yes Yes Yes! Teach it Erik - - - - - You are finally agreeing with me after 5 years!

I would think that DoF doesn't make a lot of sense when looking at actual pixels at an image.

Exactly! That is why I smile when my buddies here try to get me to start cropping in and comapring to FF and APSC (which I have done a gazillion times).

I would also suggest that looking at actual pixel on a small pitch monitor further blurs the context of DoF.

Looking at GFX res on my Dell Pro 227 ppi 32-inch 6K? It makes me want more DOF because the pixels that are in focus sear your eyeballs and the ones that aren't disapoint me.

Best regards

Erik

 

[Greg7579]

I would say that if you could (somehow) increase the per pixel sharpness of the in focus pixels, that would draw more attention to the out of focus pixels and result in less visible depth of field. This is equivalent to using a tighter CoC number in your calculator.

To get the calculated dof to agree with your subjective sense, you have to use an appropriate CoC value in your calculator, as Jim has patiently explained about to you about 10 thousand times. It doesn't mean there is something exceptional about MF sized sensors, it means you are using the wrong CoC value in your calculator. Use the right CoC value and, hey presto!, calculator and Mark I eyeballs agree.

The concept of DoF is based on acceptable level of unsharpness.

If we except maximum sharpness, DoF will be essentially nil.

On a system with sharp lenses, the zone of maximum sharpness will be thinner that it would be with blurry lenses.

I would think that DoF doesn't make a lot of sense when looking at actual pixels at an image. I would also suggest that looking at actual pixel on a small pitch monitor further blurs the context of DoF.

Best regards

Erik

What effect does using different monitors have? When I changed from 1440P to 4k, it seemed to me initially when viewing an image so the full image filled the screen, that the 4k screen made it look marginally crisper. I've got used to it now and images appear to look the same as they always did. But I've since read people say that higher resolution screens actually mask sharpness and detail. A bit confused about what is meant by this and what is correct here.

Don't overthink it. Let me makes this fact clear to everyone:

-- 4K Pro 32-inch IPS or mini-LED monitors? Your GFX image will look amazingly better than on anything else with es res (and other screen qualities I could get into that a good 4K pro monitor will have.

It looks better at full res where you only see about 60% of the image and it looks better full screen fit. Doesn't matter. Don't believe it that when you zoom to full that it will be the same as when you zoom to full res on a lesser screen. It just looks better no matter what you do. A lot better. Way way better.

-- 6K? Please.... Makes me want 8K even more. It looks way better on 6K than 4K. Trust me.

I can only dream about what a MF GFX / Hassy file is going to look like on the new 32- inch ASUS 8K mini-LED monster that will probably cost 8 or 9 Grand. I won't know because I won't buy it at that price.

 

[Greg7579]

.... the larger format comes to advantage at near optimum apertures, with lens quality probably affecting the results a lot.

Are you are concluding that for us to see any real image fidelity advantage with our MF over FF, that you have to be at optimal aperture with our GF lenses? If so, I seriously disagree with you.

I'm not being a smarty pants, Erik, I'm serious when I say that when you and Jim start talking about the mechanics of your tests and display the graphs and charts, I don't look at that much, but I am very interested in your conclusions (from not only your tests, but you experience over decades of shooting).

Stopping down for equivalent DoF to small apertures, diffraction will level the playing field.

I don't think so. I don't agree.

I do. In fact, Erik's statement is provable.

That is not what I have seen peeping at thousands of GFX vs high-res FF shots that I took. I consider each one of those shots a "test" in the field in a wide variety of situations at stuff people really shoot at. I'm not belittling your tests. I am telling you about my tests.

Well, maybe I misunderstand the nuances of his conclusion. If it is true, maybe it is shots that are so blurred with diffraction, so it doesn't matter.

I have this unprovable theory about GFX shooting that I think makes sense. Chris D and I used to talk about it a lot and I have talked about it in person with other GFX photographers and some Fuji reps....

I think it is why I think I see more than one stop difference on DOF w FF and 2 stops w APSC. Way more. It is why DOF is such a big deal with GFX, and also motion blurred pixels from insufficient handheld speed (prior to IBIS) - speed that would be OK with FF.

why don’t you share a few pairs of images where you see this difference in DoF and motion blur at equivalent settings or whatever way you prefer to show it?

And perhaps state your theory clearly, although I suspect that might not be possible?

Hey Jonas, you have popped in on me before. I tell you what.... Tell me what you shoot and then I'll explain my theory to you. It will help me understand what equipment you use and what kind of files you are used to in post. I believe all modern cameras are very good and can take great pictures, so I don't ask that to establish a ranking order.

Sony a1

Well, that is one of the best cameras in the world.

So, when you shoot it at city scenes or various landscapes (not portraits or studio) and are seeking some good DOF, do you ever get surprised in post (when viewing on a good 4K pro monitor at full res) by a certain scene at a certain aperture that seems outside the norm in your shooting experience?

Do you ever shoot at F4, 5.6, 7.1, 8 or 11 and are surprised that you have so much, or maybe less than expected DOF in different situations, lighting or sets?

Or do you think every shot is going to have a viewing result exactly in line with what a DOF table would say with the proper CoC entered and using a laser range finder for distance? Do you think there are anomalies or that different scenes might lead to a different perception about what is acceptable DOF for that shot?

Do you think moving from 50 to 100MP on a bigger sensor with great glass and viewing on a 6K monitor might have some impact on what you see concerning OOF pixels or pixels blurred by camera shake and the impact that has on the whole image and how it looks?

I don't consider a long list of questionst as a clearly stated theori.

You need to explain why MF cameras deviate more than smaller formats from that predicted by DoF calculators. This is your central claim at odds with well developed and understood theori.

Are you aware that DoF calculations are based on a predetermined image size and viewing conditions? Your frequent reference to the wonders of pro monitors lead me to think that you possibly get tricked by pixel peeping in a non-standardized way.

Perhaps I get "tricked" on a top-end pro 4k or 6k monitor. I don't know. Maybe. But I'm looking at my 60 MP Leica FF shots on that same monitor as the GFX shots, so I doubt it.

But maybe that is a point to be considered.

And I'm not denying the accuracy of the tables with proper CoC and distance. It's a great primer and indicator. I'm not sure it nails exactly what various photographers might find acceptable on various monitors, but that is OK.

Sometimes I think we underplay the value of (or the impact of) monitor technology in this digital camera world.

Are we going to start adjusting CoC for various qualities of monitors and how far our eyeballs are from the screen and maybe a variable for how tolerant we are for OOF pixels and at what point we consider it to be in focus? One thing that bothers me about CoC is you can make it anything you want to try to get something that makes sense.

But forget that. I'm not knocking CoC. It's an important academic concept in making sense of the digital madness with all of this new equipment that is no longer film driven.

--
Greg Johnson, San Antonio, Texas
https://www.flickr.com/photos/139148982@N02/albums

 

[paulraphael]

But forget that. I'm not knocking CoC. It's an important academic concept in making sense of the digital madness with all of this new equipment that is no longer film driven.

It's just a way to quantify an amount of blur. If you figure out how much blur you will tolerate for given image use (doesn't matter if it's film or digital) you can attach a number to that.

Even if you don't do it quantitatively, it's helpful to understand the concept. This will help you do a more predictable job when you're working intuitively.

If you understand the relationships affecting CoC, you'll understand why you'll be able to tolerate twice as big a blur circle on a 200ppi monitor as on a 100ppi monitor. If we're talking pixel peeping.

 

[Greg7579]

But forget that. I'm not knocking CoC. It's an important academic concept in making sense of the digital madness with all of this new equipment that is no longer film driven.

It's just a way to quantify an amount of blur. If you figure out how much blur you will tolerate for given image use (doesn't matter if it's film or digital) you can attach a number to that.

Even if you don't do it quantitatively, it's helpful to understand the concept. This will help you do a more predictable job when you're working intuitively.

If you understand the relationships affecting CoC, you'll understand why you'll be able to tolerate twice as big a blur circle on a 200ppi monitor as on a 100ppi monitor. If we're talking pixel peeping.

Yes, I get that. Thanks. It's a good and interesting concept. Pixel peeping? I don't think looking at our high-res files at full res is peeping. It's just looking. But yes, even on y 6K monitor you go to full res, and you are only looking at part of the image, so I guess it could be described as peeping. Still - I explore every image I take at full res. It's part of the fun and is sort of amazing. It's like having a super-spy camera.

 

[DavidMillier]

But forget that. I'm not knocking CoC. It's an important academic concept in making sense of the digital madness with all of this new equipment that is no longer film driven.

It's just a way to quantify an amount of blur. If you figure out how much blur you will tolerate for given image use (doesn't matter if it's film or digital) you can attach a number to that.

Even if you don't do it quantitatively, it's helpful to understand the concept. This will help you do a more predictable job when you're working intuitively.

If you understand the relationships affecting CoC, you'll understand why you'll be able to tolerate twice as big a blur circle on a 200ppi monitor as on a 100ppi monitor. If we're talking pixel peeping.

Yes, I get that. Thanks. It's a good and interesting concept. Pixel peeping? I don't think looking at our high-res files at full res is peeping. It's just looking. But yes, even on y 6K monitor you go to full res, and you are only looking at part of the image, so I guess it could be described as peeping. Still - I explore every image I take at full res. It's part of the fun and is sort of amazing. It's like having a super-spy camera.

Pixel peeping is a colloquial, sometimes slightly pejorative term, for examining images under (excessive) high magnification.

I think it is pejorative because pixel peeping people (triple alliteration!) are considered to be using modern imaging software to examine images more for their pixel quality, than for the message the whole image communicates. Kind of missing the point of the image.

But I do understand that many people do enjoy checking image quality and exploring the minutiae of the image - even if I do think it is missing the point of image making, at least for all the reasons I was ever motivated to take up photography. But everyone is different and it takes all sorts etc.

 

[Erik Kaffehr]

I would say that if you could (somehow) increase the per pixel sharpness of the in focus pixels, that would draw more attention to the out of focus pixels and result in less visible depth of field. This is equivalent to using a tighter CoC number in your calculator.

To get the calculated dof to agree with your subjective sense, you have to use an appropriate CoC value in your calculator, as Jim has patiently explained about to you about 10 thousand times. It doesn't mean there is something exceptional about MF sized sensors, it means you are using the wrong CoC value in your calculator. Use the right CoC value and, hey presto!, calculator and Mark I eyeballs agree.

The concept of DoF is based on acceptable level of unsharpness.

If we except maximum sharpness, DoF will be essentially nil.

On a system with sharp lenses, the zone of maximum sharpness will be thinner that it would be with blurry lenses.

I would think that DoF doesn't make a lot of sense when looking at actual pixels at an image. I would also suggest that looking at actual pixel on a small pitch monitor further blurs the context of DoF.

Best regards

Erik

What effect does using different monitors have? When I changed from 1440P to 4k, it seemed to me initially when viewing an image so the full image filled the screen, that the 4k screen made it look marginally crisper. I've got used to it now and images appear to look the same as they always did. But I've since read people say that higher resolution screens actually mask sharpness and detail. A bit confused about what is meant by this and what is correct here.

It depends on your viewing.

Seeing the full image, a higher resolution monitor may yield cleaner detail.

Seeing the actual pixels, a higher resolution monitor has lower magnification, thus it will show less deatil. That may the image look sharper.

Best regards

Erik

 

[Greg7579]

Well yes, in a way.... I would not explain it that way. It is a bit misleading. If you had a 18k 32 inch monitor, the whole GFX would show on screen at full res.

Depending on other tech factors like PPI, type of backlighting, HDR, nits, black levels, etc.... You would have to zoom in past full res to see more detail on screen. But the full screen full res image would look amazing...

 

[paulraphael]

Yes, I get that. Thanks. It's a good and interesting concept. Pixel peeping? I don't think looking at our high-res files at full res is peeping. It's just looking. But yes, even on y 6K monitor you go to full res, and you are only looking at part of the image, so I guess it could be described as peeping. Still - I explore every image I take at full res. It's part of the fun and is sort of amazing. It's like having a super-spy camera.

Pixel peeping is a colloquial, sometimes slightly pejorative term, for examining images under (excessive) high magnification.

I think it is pejorative because pixel peeping people (triple alliteration!) are considered to be using modern imaging software to examine images more for their pixel quality, than for the message the whole image communicates. Kind of missing the point of the image.

But I do understand that many people do enjoy checking image quality and exploring the minutiae of the image - even if I do think it is missing the point of image making, at least for all the reasons I was ever motivated to take up photography. But everyone is different and it takes all sorts etc.

I proudly peep my pixels. But it's not my end goal. I do it 1) because it's fun and 2) it helps me make certain decisions when image editing, especially for printing.

I do use the term to (gently) make fun of Greg.

 

[DavidMillier]

Yes, I get that. Thanks. It's a good and interesting concept. Pixel peeping? I don't think looking at our high-res files at full res is peeping. It's just looking. But yes, even on y 6K monitor you go to full res, and you are only looking at part of the image, so I guess it could be described as peeping. Still - I explore every image I take at full res. It's part of the fun and is sort of amazing. It's like having a super-spy camera.

Pixel peeping is a colloquial, sometimes slightly pejorative term, for examining images under (excessive) high magnification.

I think it is pejorative because pixel peeping people (triple alliteration!) are considered to be using modern imaging software to examine images more for their pixel quality, than for the message the whole image communicates. Kind of missing the point of the image.

But I do understand that many people do enjoy checking image quality and exploring the minutiae of the image - even if I do think it is missing the point of image making, at least for all the reasons I was ever motivated to take up photography. But everyone is different and it takes all sorts etc.

I proudly peep my pixels. But it's not my end goal. I do it 1) because it's fun and 2) it helps me make certain decisions when image editing, especially for printing.

I do use the term to (gently) make fun of Greg.

Let's go with proudly pixel peeping people. Quad alliteration. Who can better that...

 

[jonas ar]

.... the larger format comes to advantage at near optimum apertures, with lens quality probably affecting the results a lot.

Are you are concluding that for us to see any real image fidelity advantage with our MF over FF, that you have to be at optimal aperture with our GF lenses? If so, I seriously disagree with you.

I'm not being a smarty pants, Erik, I'm serious when I say that when you and Jim start talking about the mechanics of your tests and display the graphs and charts, I don't look at that much, but I am very interested in your conclusions (from not only your tests, but you experience over decades of shooting).

Stopping down for equivalent DoF to small apertures, diffraction will level the playing field.

I don't think so. I don't agree.

I do. In fact, Erik's statement is provable.

That is not what I have seen peeping at thousands of GFX vs high-res FF shots that I took. I consider each one of those shots a "test" in the field in a wide variety of situations at stuff people really shoot at. I'm not belittling your tests. I am telling you about my tests.

Well, maybe I misunderstand the nuances of his conclusion. If it is true, maybe it is shots that are so blurred with diffraction, so it doesn't matter.

I have this unprovable theory about GFX shooting that I think makes sense. Chris D and I used to talk about it a lot and I have talked about it in person with other GFX photographers and some Fuji reps....

I think it is why I think I see more than one stop difference on DOF w FF and 2 stops w APSC. Way more. It is why DOF is such a big deal with GFX, and also motion blurred pixels from insufficient handheld speed (prior to IBIS) - speed that would be OK with FF.

why don’t you share a few pairs of images where you see this difference in DoF and motion blur at equivalent settings or whatever way you prefer to show it?

And perhaps state your theory clearly, although I suspect that might not be possible?

Hey Jonas, you have popped in on me before. I tell you what.... Tell me what you shoot and then I'll explain my theory to you. It will help me understand what equipment you use and what kind of files you are used to in post. I believe all modern cameras are very good and can take great pictures, so I don't ask that to establish a ranking order.

Sony a1

Well, that is one of the best cameras in the world.

So, when you shoot it at city scenes or various landscapes (not portraits or studio) and are seeking some good DOF, do you ever get surprised in post (when viewing on a good 4K pro monitor at full res) by a certain scene at a certain aperture that seems outside the norm in your shooting experience?

Do you ever shoot at F4, 5.6, 7.1, 8 or 11 and are surprised that you have so much, or maybe less than expected DOF in different situations, lighting or sets?

Or do you think every shot is going to have a viewing result exactly in line with what a DOF table would say with the proper CoC entered and using a laser range finder for distance? Do you think there are anomalies or that different scenes might lead to a different perception about what is acceptable DOF for that shot?

Do you think moving from 50 to 100MP on a bigger sensor with great glass and viewing on a 6K monitor might have some impact on what you see concerning OOF pixels or pixels blurred by camera shake and the impact that has on the whole image and how it looks?

I don't consider a long list of questionst as a clearly stated theori.

You need to explain why MF cameras deviate more than smaller formats from that predicted by DoF calculators. This is your central claim at odds with well developed and understood theori.

Are you aware that DoF calculations are based on a predetermined image size and viewing conditions? Your frequent reference to the wonders of pro monitors lead me to think that you possibly get tricked by pixel peeping in a non-standardized way.

Perhaps I get "tricked" on a top-end pro 4k or 6k monitor. I don't know. Maybe. But I'm looking at my 60 MP Leica FF shots on that same monitor as the GFX shots, so I doubt it.

But maybe that is a point to be considered.

If you zoom to 100% on images from those cameras to judge DoF, we have might have explained some of your observations

And I'm not denying the accuracy of the tables with proper CoC and distance. It's a great primer and indicator. I'm not sure it nails exactly what various photographers might find acceptable on various monitors, but that is OK.

If you insist that there is something special about MF DoF, then you are in fact denying the tables

Sometimes I think we underplay the value of (or the impact of) monitor technology in this digital camera world.

Are we going to start adjusting CoC for various qualities of monitors and how far our eyeballs are from the screen and maybe a variable for how tolerant we are for OOF pixels and at what point we consider it to be in focus? One thing that bothers me about CoC is you can make it anything you want to try to get something that makes sense.

you need to adjust the CoC for the intended display size and viewing distance.