Is DOF becoming "thinner"?

[fferreres]

Many years ago, if you had a digital camera with 6MP, and using a wider aperture, the area that'd be seen in focus would be larger. Today's high resolution sensors, Pixel Shift, etc. allow much mroe detailed images, but the OOF transition is very noticeable. of course, you can always reduce resolution. But when you may want to use that resolution, you may find that you'd maybe had been better with a larger area in sharp focus.

Have you considered this factor? Is there a calculator based on "discernible DOF" that takes into account resolution? Would it be sensible as resolution increases, to maybe want to use a bit wider DOF that we where used to in the past? While the title is obviously answerable with a "no", in practice, I think this is something that has made me consider extending DOF a bit more than I was used in the past, especially, if I consider that I now also have more chance to crop as needed.

 

[ghohan422]

Many years ago, if you had a digital camera with 6MP, and using a wider aperture, the area that'd be seen in focus would be larger. Today's high resolution sensors, Pixel Shift, etc. allow much mroe detailed images, but the OOF transition is very noticeable. of course, you can always reduce resolution. But when you may want to use that resolution, you may find that you'd maybe had been better with a larger area in sharp focus.

Have you considered this factor? Is there a calculator based on "discernible DOF" that takes into account resolution? Would it be sensible as resolution increases, to maybe want to use a bit wider DOF that we where used to in the past? While the title is obviously answerable with a "no", in practice, I think this is something that has made me consider extending DOF a bit more than I was used in the past, especially, if I consider that I now also have more chance to crop as needed.

All DOF calculators have to have some idea of what is "in focus" and "out of focus" based on some acceptable level of sharpness. They use the idea of a "Circle of Confusion" which for digital cameras you can think of as how much lens blur there is before the blur is spreading across multiple pixels. More pixels on newer cameras means that pixel level sharpness is possible over a smaller region in front of and in back of your subject.

Some DOF calculators expose this to you in a way that makes sense. I'm not going to go looking for the best one now but you can try this one here:

https://www.cambridgeincolour.com/tutorials/dof-calculator.htm

If you click "Show Advanced" you can change the max print dimension. Increasing this is equivalent to going up to a camera with more megapixels and you see that it will tell you that the DOF is thinner at larger print dimensions. It's really the same blur in terms of percentage of the image but there is more blur per pixel in the image.

 

[fferreres]

Many years ago, if you had a digital camera with 6MP, and using a wider aperture, the area that'd be seen in focus would be larger. Today's high resolution sensors, Pixel Shift, etc. allow much mroe detailed images, but the OOF transition is very noticeable. of course, you can always reduce resolution. But when you may want to use that resolution, you may find that you'd maybe had been better with a larger area in sharp focus.

Have you considered this factor? Is there a calculator based on "discernible DOF" that takes into account resolution? Would it be sensible as resolution increases, to maybe want to use a bit wider DOF that we where used to in the past? While the title is obviously answerable with a "no", in practice, I think this is something that has made me consider extending DOF a bit more than I was used in the past, especially, if I consider that I now also have more chance to crop as needed.

All DOF calculators have to have some idea of what is "in focus" and "out of focus" based on some acceptable level of sharpness. They use the idea of a "Circle of Confusion" which for digital cameras you can think of as how much lens blur there is before the blur is spreading across multiple pixels. More pixels on newer cameras means that pixel level sharpness is possible over a smaller region in front of and in back of your subject.

Some DOF calculators expose this to you in a way that makes sense. I'm not going to go looking for the best one now but you can try this one here:

https://www.cambridgeincolour.com/tutorials/dof-calculator.htm

If you click "Show Advanced" you can change the max print dimension. Increasing this is equivalent to going up to a camera with more megapixels and you see that it will tell you that the DOF is thinner at larger print dimensions. It's really the same blur in terms of percentage of the image but there is more blur per pixel in the image.

Yes, good point. How does one account for that? With increased resolution, it may now be best to sometimes use f2.8 where one would have used maybe 2.0, etc. Is there a calculator that could recommend an adjusted DOF aware of resolution?

 

[drj3]

With more pixels, you are probably enlarging the image more and thus decreasing the DOF. The DOF calculators assume a specific size image viewed at a specific distance with what looks relatively sharp (for someone with less than 20/20 vision). For example the Cambridge DOF calculator assumes a 10 inch maximum print dimension viewed at 25 cm (or 10 inches). The Cambridge DOF calculator Advanced tab gives you the advantage of changing the manufacturer default to 20/20 vision.

The attached gives you the flexibility to determine the DOF based specific situations.

Loading…

www.photopills.com

 

[Qui Tam]

I know some people whose lives and shots have been ruined by the ever-thinning DOF and ensuing "fixes" such as focus stacking, which can introduce more problems than they solve.

It's not a rabbit hole worth descending.

 

[WunWegWunDarWun]

Many years ago, if you had a digital camera with 6MP, and using a wider aperture, the area that'd be seen in focus would be larger. Today's high resolution sensors, Pixel Shift, etc. allow much mroe detailed images, but the OOF transition is very noticeable. of course, you can always reduce resolution. But when you may want to use that resolution, you may find that you'd maybe had been better with a larger area in sharp focus.

Have you considered this factor? Is there a calculator based on "discernible DOF" that takes into account resolution? Would it be sensible as resolution increases, to maybe want to use a bit wider DOF that we where used to in the past? While the title is obviously answerable with a "no", in practice, I think this is something that has made me consider extending DOF a bit more than I was used in the past, especially, if I consider that I now also have more chance to crop as needed.

Another way to look at it is that focus is becoming more focused

 

[Leonard Migliore]

Many years ago, if you had a digital camera with 6MP, and using a wider aperture, the area that'd be seen in focus would be larger. Today's high resolution sensors, Pixel Shift, etc. allow much mroe detailed images, but the OOF transition is very noticeable. of course, you can always reduce resolution. But when you may want to use that resolution, you may find that you'd maybe had been better with a larger area in sharp focus.

Have you considered this factor? Is there a calculator based on "discernible DOF" that takes into account resolution? Would it be sensible as resolution increases, to maybe want to use a bit wider DOF that we where used to in the past? While the title is obviously answerable with a "no", in practice, I think this is something that has made me consider extending DOF a bit more than I was used in the past, especially, if I consider that I now also have more chance to crop as needed.

What you're looking for is the circle of confusion. This is the diameter of blur that is used to calculate depth of field. The standard circle of confusion is based on the alleged ability of a person with "normal" vision to see blur on an 8" X 10" print; this is typically considered to be 0.25mm for a viewing distance of 250mm. So if you have a full-frame sensor, you need to magnify the image 8 times to generate an 8" X 10" print. The circle of confusion for a full-frame sensor is then 0.25mm /8 or about 30 microns.

DoF calculators such as DOFMaster allow you to enter any desired circle of confusion. It defaults to the value I mentioned above but if you want to print bigger, you can put in a smaller number and the program will calculate the depth of field accordingly.

 

[fferreres]

Thanks everyone. Great answers, although I am still a bit wondering in more pragmatical ways looking for a rule of thumb, I noted some of the responses and tools. Whenever I can, I think I will also add DOF unless really needed. It's been a few times where part of the subject is clearly not in focus the moment I need to crop. And this would get worst if I doubled the resolution to 45MP or worst using Pixel Shift and printing or large hugher res display (4k or 8k projector)

 

[Leonard Migliore]

Thanks everyone. Great answers, although I am still a bit wondering in more pragmatical ways looking for a rule of thumb, I noted some of the responses and tools. Whenever I can, I think I will also add DOF unless really needed. It's been a few times where part of the subject is clearly not in focus the moment I need to crop. And this would get worst if I doubled the resolution to 45MP or worst using Pixel Shift and printing or large hugher res display (4k or 8k projector)

If you want trouble, try shooting macro. The first rule of macro is, whatever your depth of field is, it's not enough.

Then, if you want to quantify things, you find that normal depth of field calculations don't work and you have to incorporate the pupil magnification, which you usually don't know.

And then, if you just want to avoid problems by stopping down, you get totally messed up by diffraction and have lots of depth of field but nothing's sharp.

So then you cover your bases by stacking focus and find that you're changing your perspective when you do that and everything looks strange. Sharp but strange.

Be glad you're working in the big world. Depth of field problems there are trivial.

--
Leonard Migliore

 

[Christof21]

Many years ago, if you had a digital camera with 6MP, and using a wider aperture, the area that'd be seen in focus would be larger. Today's high resolution sensors, Pixel Shift, etc. allow much mroe detailed images, but the OOF transition is very noticeable. of course, you can always reduce resolution. But when you may want to use that resolution, you may find that you'd maybe had been better with a larger area in sharp focus.

Have you considered this factor? Is there a calculator based on "discernible DOF" that takes into account resolution? Would it be sensible as resolution increases, to maybe want to use a bit wider DOF that we where used to in the past? While the title is obviously answerable with a "no", in practice, I think this is something that has made me consider extending DOF a bit more than I was used in the past, especially, if I consider that I now also have more chance to crop as needed.

Higher resolution has absolutely nothing to do with dof.

At "normal" distance, the eyes can not see the difference anymore. If you get closer, either you see pixelisation either there is more resolution and you may consider it oof, but more pixels is always better !!!

Imagine your image has infinite resolution, do you think the dof suddenly gets thinner. Nonsense !

Do not hesitate to get more resolution, it won't change dof at all.

 

[Leonard Migliore]

Many years ago, if you had a digital camera with 6MP, and using a wider aperture, the area that'd be seen in focus would be larger. Today's high resolution sensors, Pixel Shift, etc. allow much mroe detailed images, but the OOF transition is very noticeable. of course, you can always reduce resolution. But when you may want to use that resolution, you may find that you'd maybe had been better with a larger area in sharp focus.

Have you considered this factor? Is there a calculator based on "discernible DOF" that takes into account resolution? Would it be sensible as resolution increases, to maybe want to use a bit wider DOF that we where used to in the past? While the title is obviously answerable with a "no", in practice, I think this is something that has made me consider extending DOF a bit more than I was used in the past, especially, if I consider that I now also have more chance to crop as needed.

Higher resolution has absolutely nothing to do with dof.

At "normal" distance, the eyes can not see the difference anymore. If you get closer, either you see pixelisation either there is more resolution and you may consider it oof, but more pixels is always better !!!

Imagine your image has infinite resolution, do you think the dof suddenly gets thinner. Nonsense !

Do not hesitate to get more resolution, it won't change dof at all.

So something that looks sharp on your phone looks sharp on a 40" X 60" print?

Not for me.

I support using tighter constraints on depth of field calculations when you expect to print large than when you just send stuff out on Instagram.

 

[Christof21]

Many years ago, if you had a digital camera with 6MP, and using a wider aperture, the area that'd be seen in focus would be larger. Today's high resolution sensors, Pixel Shift, etc. allow much mroe detailed images, but the OOF transition is very noticeable. of course, you can always reduce resolution. But when you may want to use that resolution, you may find that you'd maybe had been better with a larger area in sharp focus.

Have you considered this factor? Is there a calculator based on "discernible DOF" that takes into account resolution? Would it be sensible as resolution increases, to maybe want to use a bit wider DOF that we where used to in the past? While the title is obviously answerable with a "no", in practice, I think this is something that has made me consider extending DOF a bit more than I was used in the past, especially, if I consider that I now also have more chance to crop as needed.

Higher resolution has absolutely nothing to do with dof.

At "normal" distance, the eyes can not see the difference anymore. If you get closer, either you see pixelisation either there is more resolution and you may consider it oof, but more pixels is always better !!!

Imagine your image has infinite resolution, do you think the dof suddenly gets thinner. Nonsense !

Do not hesitate to get more resolution, it won't change dof at all.

So something that looks sharp on your phone looks sharp on a 40" X 60" print?

Not for me.

1/ More resolution does not mean bigger print ! This is nonsense. More resolution means more resolution, period.

I have aps-c. If I go FF, this does not mean at all that I will print bigger. This means for me mainly more possibilities for cropping if needed. In this case, this is like shooting with a longer FL and a larger f#

2/ do you think movies are recorded once for tv and once for theater ? Considering the difference of display size, this should be done.. the answer is off course no (or not big change)

The difference with a phone is bigger because the eyes can not see an image too close. So the angle you view the image is much smaller.

I support using tighter constraints on depth of field calculations when you expect to print large

Then simply consider the size of the print, not the resolution.

than when you just send stuff out on Instagram.

 

[Christof21]

Thanks everyone. Great answers, although I am still a bit wondering in more pragmatical ways looking for a rule of thumb, I noted some of the responses and tools. Whenever I can, I think I will also add DOF unless really needed. It's been a few times where part of the subject is clearly not in focus the moment I need to crop.

If you crop, simply consider a smaller circle of confusion

And this would get worst if I doubled the resolution to 45MP or worst using Pixel Shift and printing or large hugher res display (4k or 8k projector)

The resolution per see has absolutely no impact on dof.

 

[fferreres]

Thanks everyone. Great answers, although I am still a bit wondering in more pragmatical ways looking for a rule of thumb, I noted some of the responses and tools. Whenever I can, I think I will also add DOF unless really needed. It's been a few times where part of the subject is clearly not in focus the moment I need to crop. And this would get worst if I doubled the resolution to 45MP or worst using Pixel Shift and printing or large hugher res display (4k or 8k projector)

If you want trouble, try shooting macro. The first rule of macro is, whatever your depth of field is, it's not enough.

Then, if you want to quantify things, you find that normal depth of field calculations don't work and you have to incorporate the pupil magnification, which you usually don't know.

And then, if you just want to avoid problems by stopping down, you get totally messed up by diffraction and have lots of depth of field but nothing's sharp.

So then you cover your bases by stacking focus and find that you're changing your perspective when you do that and everything looks strange. Sharp but strange.

Be glad you're working in the big world. Depth of field problems there are trivial.

I know. It cost me a lot to try to get this little animal in focus. Actually, I went through all the issues you mentioned, made worst because this lens isn't supposed to be very "macroable" I guess (quickly loses harpness or becomes razor thin).

Actually, instead of just trying to fiddle with the camera, I had to fiddle with myself and position so I was mostly paralel to the beast, by milliliters difference. Anything above f8 was totally diffracted. DOF may be 2 mm here.


CZ 50/1.4 QBM with Helicoid @ ˜f8

I guess regarding DOF, I will just try to use a bit longer DOF. I find that f1.4 and f2, for example, mean little change. Yes, if I want/need f1.4 ...great, but it also sacrifices a lof of image quality, corner performance, lowered contrast (regardless of which lens).

 

[fferreres]

Many years ago, if you had a digital camera with 6MP, and using a wider aperture, the area that'd be seen in focus would be larger. Today's high resolution sensors, Pixel Shift, etc. allow much mroe detailed images, but the OOF transition is very noticeable. of course, you can always reduce resolution. But when you may want to use that resolution, you may find that you'd maybe had been better with a larger area in sharp focus.

Have you considered this factor? Is there a calculator based on "discernible DOF" that takes into account resolution? Would it be sensible as resolution increases, to maybe want to use a bit wider DOF that we where used to in the past? While the title is obviously answerable with a "no", in practice, I think this is something that has made me consider extending DOF a bit more than I was used in the past, especially, if I consider that I now also have more chance to crop as needed.

Higher resolution has absolutely nothing to do with dof.

I know.

At "normal" distance, the eyes can not see the difference anymore. If you get closer, either you see pixelisation either there is more resolution and you may consider it oof, but more pixels is always better !!!

The sharper the lens I get, the more I notice it, especially if print is larger. I did not experience this with my 12MP 5D Classic.

Imagine your image has infinite resolution, do you think the dof suddenly gets thinner. Nonsense !

Infinite resolution may mean I get to inspect closer without any pixelation. I am limited by my eye and the image distance from myself. WIth large MP cameras and sharp lenses, if I look in a large screen, I see OOF areas that I wanted ideally to not be blurrier than other parts. That's my point...that I now thing about increasing DOF more without sacrificing much bokeh. This is contrary to when one starts trying to get ever thiner DOFs. I am pivoting back to higher DOF because I do notice some hairs semi blurred and other ultra sharp and it bother me personally (unless intended).

Do not hesitate to get more resolution, it won't change dof at all.

:-( I think I probably didn't explain the point correctly, which I notice more how the nose, some hairs (at almost the same distance, maybe 1 cm apart) are rendered one sharp the other not at all. And I intended all the subject to be equally nitid (I didn't, with a lower res camera). Also, displays of 4k (eg. 160 diagonal projector) that are seen from 10 inch distance, make this more obvious.

 

[fferreres]

Many years ago, if you had a digital camera with 6MP, and using a wider aperture, the area that'd be seen in focus would be larger. Today's high resolution sensors, Pixel Shift, etc. allow much mroe detailed images, but the OOF transition is very noticeable. of course, you can always reduce resolution. But when you may want to use that resolution, you may find that you'd maybe had been better with a larger area in sharp focus.

Have you considered this factor? Is there a calculator based on "discernible DOF" that takes into account resolution? Would it be sensible as resolution increases, to maybe want to use a bit wider DOF that we where used to in the past? While the title is obviously answerable with a "no", in practice, I think this is something that has made me consider extending DOF a bit more than I was used in the past, especially, if I consider that I now also have more chance to crop as needed.

Higher resolution has absolutely nothing to do with dof.

At "normal" distance, the eyes can not see the difference anymore. If you get closer, either you see pixelisation either there is more resolution and you may consider it oof, but more pixels is always better !!!

Imagine your image has infinite resolution, do you think the dof suddenly gets thinner. Nonsense !

Do not hesitate to get more resolution, it won't change dof at all.

So something that looks sharp on your phone looks sharp on a 40" X 60" print?

Not for me.

I support using tighter constraints on depth of field calculations when you expect to print large than when you just send stuff out on Instagram.

+1 not for me! For Instagram, I couldn't care less. But Instagram is more like a thumbnail.

 

[fferreres]

Many years ago, if you had a digital camera with 6MP, and using a wider aperture, the area that'd be seen in focus would be larger. Today's high resolution sensors, Pixel Shift, etc. allow much mroe detailed images, but the OOF transition is very noticeable. of course, you can always reduce resolution. But when you may want to use that resolution, you may find that you'd maybe had been better with a larger area in sharp focus.

Have you considered this factor? Is there a calculator based on "discernible DOF" that takes into account resolution? Would it be sensible as resolution increases, to maybe want to use a bit wider DOF that we where used to in the past? While the title is obviously answerable with a "no", in practice, I think this is something that has made me consider extending DOF a bit more than I was used in the past, especially, if I consider that I now also have more chance to crop as needed.

Higher resolution has absolutely nothing to do with dof.

At "normal" distance, the eyes can not see the difference anymore. If you get closer, either you see pixelisation either there is more resolution and you may consider it oof, but more pixels is always better !!!

Imagine your image has infinite resolution, do you think the dof suddenly gets thinner. Nonsense !

Do not hesitate to get more resolution, it won't change dof at all.

So something that looks sharp on your phone looks sharp on a 40" X 60" print?

Not for me.

1/ More resolution does not mean bigger print ! This is nonsense. More resolution means more resolution, period.

The point is not to lower resolution. But once you have access to higher resolution capture, and higher resolution displays (or print) you ma find yourself having wanted to have used a bigger DOF to start with. This is the main point. And if you crop, this makes it obviously more evident (in the past, you couldn't even crop).

So ...I am not complaining about the resolution. Just observing I am stopping down a bit more now, than before, for this reason.

 

[Christof21]

Many years ago, if you had a digital camera with 6MP, and using a wider aperture, the area that'd be seen in focus would be larger. Today's high resolution sensors, Pixel Shift, etc. allow much mroe detailed images, but the OOF transition is very noticeable. of course, you can always reduce resolution. But when you may want to use that resolution, you may find that you'd maybe had been better with a larger area in sharp focus.

Have you considered this factor? Is there a calculator based on "discernible DOF" that takes into account resolution? Would it be sensible as resolution increases, to maybe want to use a bit wider DOF that we where used to in the past? While the title is obviously answerable with a "no", in practice, I think this is something that has made me consider extending DOF a bit more than I was used in the past, especially, if I consider that I now also have more chance to crop as needed.

Higher resolution has absolutely nothing to do with dof.

At "normal" distance, the eyes can not see the difference anymore. If you get closer, either you see pixelisation either there is more resolution and you may consider it oof, but more pixels is always better !!!

Imagine your image has infinite resolution, do you think the dof suddenly gets thinner. Nonsense !

Do not hesitate to get more resolution, it won't change dof at all.

So something that looks sharp on your phone looks sharp on a 40" X 60" print?

Not for me.

1/ More resolution does not mean bigger print ! This is nonsense. More resolution means more resolution, period.

The point is not to lower resolution. But once you have access to higher resolution capture, and higher resolution displays (or print) you ma find yourself having wanted to have used a bigger DOF to start with. This is the main point.

Ok, then this is a different question.

Higher resolution is always better. Extract from what you said:

".And this would get worst if I doubled the resolution to 45MP"

It does not get worse. I understand your point now but it needed this clarification.

And if you crop, this makes it obviously more evident (in the past, you couldn't even crop).

If you crop, the input for the dof calculator changes.

I think the best way to avoid errors it to use it this way: if you crop 1.2 times, change the FL (multiply by 1.2) and f# (multiply by 1.2). Changing only the circle of confusion will not work or the formula to change it is not simple.

So ...I am not complaining about the resolution

Good, we agree then

. Just observing I am stopping down a bit more now, than before, for this reason.

There is only one focus plane... when you zoom/crop, depending on the crop factor, you will see OOF parts more visible.

But it is true you can have all in focus with a circle of confusion which has the size of a pixel... please note that a dof calculator does not take into account diffraction so this will not be correct most of the time. Photography is a compromise, there are applications (very very few, hard to find) which will help you find the best compromise, but I would not bother about it.

Just choose a circle of confusion which meets better your own criterias. I think that choosing a smaller one (like 1.5 times smaller) is a good thing

 

[threw the lens]

This thread is comical.

All it is saying is that you expect images to be sharp and fast lenses are supposed to be so good but funnily enough if you use them wide open not all the photo is sharp- !!!

Well blow me down.

I suggest you go over to a m4/3 ILC system with the kit f5.6 lens if you don't get it. It will be potentially sharp all over, the kit lens is sharp, I have to give it its due. Leave the fast lenses on the big systems to people with the big ideas- even if they are fooling themselves.

 

[Gerry Winterbourne]

Many years ago, if you had a digital camera with 6MP, and using a wider aperture, the area that'd be seen in focus would be larger. Today's high resolution sensors, Pixel Shift, etc. allow much more detailed images, but the OOF transition is very noticeable.

Worry 1. I don't understand this. The transition from being in focus to out of focus is a gradual one based on the geometry of distance from the focus plane. It doesn't go in steps so I don't see how it can be more or less noticeable just because the pixel density changes.

Just to be clear on this, I assume that you are looking at all pictures at the same physical size. If you are looking at 100% then the physical size of the image changes and DOF changes with it; so the transition won't be in the same place. As a 100% image on a high-MP sensor is bigger than on a low-MP sensor you enlarge the details more so everything becomes more defined, not just the transition.

Worry 2. You seem to be talking about comparison between low-MP and high-MP images. But unless you are comparing two photos taken effectively at the same time from the same place with the same lens and exposure settings and viewed in exactly the same way there isn't actually anything to compare. Without that all you are doing is comparing vague impressions.

of course, you can always reduce resolution. But when you may want to use that resolution, you may find that you'd maybe had been better with a larger area in sharp focus.

Have you considered this factor? Is there a calculator based on "discernible DOF" that takes into account resolution?

It's easy enough to do this with any DOF calculator that allows you to choose your own value of circle of confusion.

First, CoC is used to define how blurred a real point looks by virtue of being away from the focus plane. You can sometimes see it called the blur circle; and this is actually a better name for it. Something that is really a point will look blurred in a photo for several reasons, even when it is at the focus distance; and standard DOF calculations ignore all of them:

Lens resolution: unless the lens is perfectly corrected its aberrations will always mean that what should be a point is actually slightly blurred.

Pixel resolution: most cameras merge the output from several sensels into a single pixel; this introduces blur.

Diffraction: all apertures cause diffraction because of their edges; the effect is trivial at wide apertures but becomes significant at small apertures.

Worry 3. As these all increase the real blur in the image it means that they all (individually and collectively) reduce the DOF you see compared to standard calculations. So using, for example, a higher-MP camera will yield a real DOF that is deeper than with a lower-MP camera (all other things being kept equal). But you seem to be talking as though the higher-MP camera reduces DOF("thinner" as in your thread title).

Each of those factors creates its own blur circle. So to take account of them you need to find out the magnitude of each and add it to the CoC to get a bigger total, blur circle.

Assuming a high quality lens you can probably ignore lens blur - whatever its size it will be the same whatever sensor it's in front of. And while there are concerns that current very high MP cameras can out-resolve lenses it's probably safe to say that lens resolution is generally less of a factor than pixel resolution. For pixel resolution it's probably reasonable to take the size of a single pixel; on FF that gives 12 microns at 6MP (if there ever was a 6MP FF camera; at 12MP the pixel size is about 8 microns), 4.4 microns at 36MP. Diffraction blur can be calculated like this https://www.photopills.com/calculators/diffraction

Note that because standard DOF calculations ignore these then whatever the calculations say, if you view an image under the exact conditions assumed in the calculations the real DOF will be a bit shallower than calculated.

Now, just to complicate things, if you add the blur circle sizes noted to the assumed CoC you get a bigger value and the standard calculations take this as the amount of blur you are happy yo accept; so the calculated DOF increases! but we want to know how much DOF decreases so we have to fool the calculator by subtracting the values.

Here's an illustration. I'll take a 50mm lens on FF at f/5.6, focused at 10m. This is less than the hyperfocal distance so we can see the results for both ends of DOF.

The standard calculation, which uses 30 microns for CoC, gives a total DOF of 24.8m (6- 30.8m). If I use the 12MP camera and assume its pixel blur is 8 microns (as above) then I change CoC to 30 --8 =22 microns. This gives a total DOF of only 13.1m (6.7 - 19.8), which is much less than the standard calculation.

Going to the 36MP sensor I assume 4 microns of pixel blur (because the DOF calculator I'm using doesn't allow fractions) so CoC goes to 30 -4 = 26 microns. That gives a total of 17.8m (6.3 - 24.1m), which is still shallower than the standard calculation but noticeably deeper than the 12MP sensor.

Would it be sensible as resolution increases, to maybe want to use a bit wider DOF that we where used to in the past?

Worry 4. This seems back to front. First, I can't see why using a different body would make me want a different DOF - what I see in the picture is what I want. If (whatever the calculations say) I was happy with the DOF given by a low MP camera then that's still what I want.

What I think you may be getting at is not how to change DOF but how to get the same DOF by altering the camera settings. To do that you go to the DOF calculator, find the actual old DOF using the appropriate modification of CoC; then change CoC again to the newer sensor value; and then adjust aperture until DOF reverts to the old value.

Using the illustration from above, if I change f-stop from f/5.6 to f/4.8 I get a total DOF of 13m, which is effectively the same as the old 13.1m value.

While the title is obviously answerable with a "no", in practice, I think this is something that has made me consider extending DOF a bit more than I was used in the past, especially, if I consider that I now also have more chance to crop as needed.

Don't forget that cropping changes everything so you'd need to start with a new sety of calculations.

--
---
Gerry
___________________________________________
First camera 1953, first Pentax 1985, first DSLR 2006
http://www.pbase.com/gerrywinterbourne
[email protected]