[DSLRnovice]

This might not be the best title for a thread. In the film days, DOF was planned with print size and a standard viewing distance in mind. The same concept was used with slide film because there is a minimum distance one can view a projected slide and still be able to see the entire mage.

The formulas I used for 35mm and Medium Format were fine for the 30x40" prints I made back then. Most lens DOF scales needed an additional stop down for my use.

However, when viewing in the camera at 16x, I begin to doubt if I have enough DOF in some areas. Do you think the ability to zoom in both in the camera and on the monitor is provoking unreasonable expectations? Another problem I have with modern lenses is that they don't always have the normal distance scale so I can use the hyperfocal distance method to get the maximum DOF.

Has anyone done any testing comparing acceptable sharpness in print enlargements vs magnification in the camera particularly?

[Keith23]

No comment on the DOF because I think it is personal preference for the appearance of the image. As for calculating DOF, I always have my phone and use an app, if I cannot judge from a test image.

[stevebrot]

However, when viewing in the camera at 16x, I begin to doubt if I have enough DOF in some areas

Magnification defeats DOF...

As you noted, the standard notion is based on perception of acceptable sharpness for an 8x10 print at about 20" viewing distance. In reality, it does not exist. There is a single point of focus and increasing blur both in front of and behind that point. If magnification is low enough the amount of blur overall approaches zero. If magnification is high enough, nothing appears sharp.

How much DOF is enough? Both the optical viewfinder and the rear LCD provide exaggerated DOF relative to an 8x10 print. Most computer monitors will show less DOF than an 8x10 print. As with @Keith23 above, I use a phone app to approximate and compare DOF, but seldom attempt fine control of DOF. Doing so is futile.

BTW...compare my avatar image to the photo below. The only difference is size (magnification).




Steve

[pschlute]

However, when viewing in the camera at 16x, I begin to doubt if I have enough DOF in some areas.

Looking at a 16x zoom on the camera LCD screen (K-1) is zooming to over 100%. You will not get an accurate representation of your picture. Stick to 10x (100%)

But for critical analysis you want to be viewing on a computer screen.

[beholder3]

In the film days, DOF was planned with print size and a standard viewing distance in mind.

It still makes sense to assume people will be watching images as a whole as if they were looking through the 43mm limited lens. Then a human can resolve about 6 MPx.

If you have techno geeks zooming into images on screen on inspecting prints with their nose to the canvas, obviously the requirements could go to infinity. Magnification is proportional to DoF/noise/dynamic range perception.

It is a very subjective personal choice if you care for those types of people's opinion or no.

It all is between the ears of the people.

[GUB]

Look I agree that all our pixel peeping is creating overkill emphasis on sharpness. But one thing I struggle with is when you check out the DofMaster calculator you find the circle of confusion applied to FF (D800 coz K1 isn't there) is .03mm.
Online Depth of Field Calculator
If you calculate the size of a K1 pixel (.005mm) then you can see that a coc blur covering almost 30 pixels is considered acceptable. I guess that we probably lost the plot in the drive for more megapixels. But personally when I am working something out I halve the coc which is about 7 pixels covered and allows for a fair bit of cropping.

[TonyW]

....The formulas I used for 35mm and Medium Format were fine for the 30x40" prints I made back then. Most lens DOF scales needed an additional stop down for my use.

Basing the needs for depth of field on interpreting from the camera lens markings and stopping down further I think is still sound practice and you may find that you get more accuracy by using charts and focusing on a specific distance (which may be estimated or even better assessed using an accurate rangefinder). All this of course assumes you have the time and inclination to go to this trouble.

However, when viewing in the camera at 16x, I begin to doubt if I have enough DOF in some areas. Do you think the ability to zoom in both in the camera and on the monitor is provoking unreasonable expectations? Another problem I have with modern lenses is that they don't always have the normal distance scale so I can use the hyperfocal distance method to get the maximum DOF.

I must say I am not 100% comfortable in viewing the LCD and relying entirely on what it shows as being dead accurate. I would tend to use it more to check focus is at a particular point rather than rely on showing an accurate portrayal of DoF.

Accurate scaling of the image is I think one of the perils of zooming in on the camera LCD (as it is with Photoshop). Further complicated by what the camera uses to preview image on LCD does it use the embedded (and if it does is the embedded full size or some other).

What I can say about monitors is that if you are viewing at 100% then you are probably looking at the image at around 3x a finished print size (Standard monitor and Canon printer) and therefore it would be reasonable to expect an inaccurate view of a print. If you are web viewing you really need to look at your image downsampled to whatever size you are presenting to the web. In both cases your data probably needs to be treated differently as far as sharpening goes

[DSLRnovice]

Looking at a 16x zoom on the camera LCD screen (K-1) is zooming to over 100%. You will not get an accurate representation of your picture. Stick to 10x (100%)

But for critical analysis you want to be viewing on a computer screen.

You gave me an idea. I could shoot a scene at different apertures, make a 11x16" print with the f-calc settings, and see which one viewed in the camera best matches the print.

---------- Post added 11-07-2017 at 09:21 AM ----------

Basing the needs for depth of field on interpreting from the camera lens markings and stopping down further I think is still sound practice and you may find that you get more accuracy by using charts and focusing on a specific distance (which may be estimated or even better assessed using an accurate rangefinder). All this of course assumes you have the time and inclination to go to this trouble.

Years ago when I was using a view camera I saw a pocket device that worked like a rangefinder camera. You would align the points for the subject you wanted in sharp focus and it would indicate the distance. I should have bought it. For even when I switched to Medium Format the bellows focusing was difficult for calculating the distance. Now I have the DFA 28-105 I have to guess the distance again.

---------- Post added 11-07-2017 at 09:33 AM ----------

Look I agree that all our pixel peeping is creating overkill emphasis on sharpness. But one thing I struggle with is when you check out the DofMaster calculator you find the circle of confusion applied to FF (D800 coz K1 isn't there) is .03mm.
Online Depth of Field Calculator
If you calculate the size of a K1 pixel (.005mm) then you can see that a coc blur covering almost 30 pixels is considered acceptable. I guess that we probably lost the plot in the drive for more megapixels. But personally when I am working something out I halve the coc which is about 7 pixels covered and allows for a fair bit of cropping.

.03 worked well with 35mm film, but it could be the case that the entire image was less sharp than with today's FF digital cameras. Without cropping and at 10x print sizes, it still might be sufficient. One thing I have found that even focused at infinity, distant objects like trees do not reproduce too crisp due to atmospheric haze, which can add to the depth perception when viewing.

Extreme DOP works best when large object like mountains are in the distance, and smaller objects are up close. There is still the advantage of the view camera movements in many cases.

[TonyW]

......Years ago when I was using a view camera I saw a pocket device that worked like a rangefinder camera. You would align the points for the subject you wanted in sharp focus and it would indicate the distance. I should have bought it. For even when I switched to Medium Format the bellows focusing was difficult for calculating the distance. Now I have the DFA 28-105 I have to guess the distance again.

You probably came across something like the Watameter
Google Image Result for http://farm3.static.flickr.com/2662/4139576367_51a47331a3.jpg
I have a couple of these and can be useful in the field where time not a problem and accuracy paramount. They do appear on ebay occasionally so might be worth adding to toolkit. There are also a few apps for Android and iOS that look interesting but no idea if they are accurate or easy/fast to use.

[GUB]

.03 worked well with 35mm film, but it could be the case that the entire image was less sharp than with today's FF digital cameras.

Quite agree but bear in mind that, as the OPs first sentence says , the standard was developed based based on a viewing distance and printed image size and the viewers ability to resolve detail rather than the ability of film to record detail. Seems our viewing requirements have changed..

[DSLRnovice]

You probably came across something like the Watameter
Google Image Result for http://farm3.static.flickr.com/2662/4139576367_51a47331a3.jpg
I have a couple of these and can be useful in the field where time not a problem and accuracy paramount. They do appear on ebay occasionally so might be worth adding to toolkit. There are also a few apps for Android and iOS that look interesting but no idea if they are accurate or easy/fast to use.

The device I found was plastic and I thought they were asking too much for it. I think it was smaller than the Watameter, which I liked, but I don't know how accurate it was.