M8 and hyperfocal distance

[aulia]

Advertisement (gone after registration)

HAPPY NEW YEAR TO ALL!

 

I have been lurking in this forum for quite some time, and have gotten lots of useful information. Today, I would like to seek the advice on an area that has been confusing me.

 

On the Leica or CV lenses, there is the markings for the hyperfocal distance by aligning the infinity mark to the appropriate colored lines of our aperture setting.

 

For the M8, there is a multiplier effect of 1.33 for focal length, my question is:

 

1. Are the markings for setting of the hyperfocal distance still correct?

2. For zone focusing, are the distance marking still correct? Or do we also multiply by the same factor?

 

:confused:

 

Thanks

 

Aulia

[andit]

Hi Aulia,

 

Welcome to the forum and a very happy new year to you too. The 1.33 Field Of View effect has no effect on the focusing distance, or the hyperfocal distance. It only has to do with what the lens takes in from left to right and top to bottom.

 

If, for instance, you put a 75mm lens on your camera, the focal length is still only 75mm (focal length is the distance from the centre of the front lens element to the film plane/sensor surface). What changes is that a full frame 35mm image would take in more from left to right and from top to bottom. The image coverage is thus similar to what a 100mm lens would have given you. However, you still only have a focal length of 75mm (but have cropped the image by 1.33).

 

Thus, your camera to subject distance (your focusing distance) remains the same, irrespective of what the FOV factor is. So, zone focus as you would with a normal 35mm film camera.

 

Hope this help you a bit.

 

Andreas

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[lct]

Standard rule of thumb is to use the nearest faster f stop of the DoF scale i.e. f/5.6 when you choose f/8 for metering for instance. This is due to the fact that the CoC value of the M8 is 1.33 times smaller than that of 35mm film.

[nhabedi]

The 1.33 Field Of View effect has no effect on the focusing distance, or the hyperfocal distance.

 

Unfortunately, this is wrong - there is an effect on the hyperfocal distance - see here for a detailed explanation. So, as lct says, the summary is that you should use the next faster f stop instead.

 

Keep in mind though, that these rules are always based on very subjective rules of thumb. The "circle of confusion" is not some natural constant, but rather an empirical value based on typical print sizes and viewing distances. If you pixel-peep at 100 percent on a computer screen, all bets are off anyway.

[andit]

Ok, sorry... I just learnt something new today. This is an interesting article, and I follow what they say.

 

That would mean when putting a 35mm back on a medium format camera (such as a Hassy 501), you loose most of your depth of field... (your image size drops from 60x60 to 24x36).

 

Andreas

[lars_bergquist]

I'm sorry, we have covered this before at least a couple of times, but here we go again:

 

The d.o.f. scale was wrong already before the M8. It goes back to the 1920's when it was assumed that the 35mm negative would be enlarged at most 3x, to give a print to compete with a contact print from a 6x9cm roll film camera. But some mischievous spoilsports, including the redoubtable Dr. Paul Wolff, soon started to do prints in 30, 40 or even 50cm sizes. And today a simple test will convince you that the computational basis of the Leica d.o.f. tables, which have not changed for eighty years, are insufficient even for a decent 10x15cm (4x6'') print.

 

Therefore experienced photographers used the 'minus one stop' rule recommended above already decades ago. So did I. Enlarging to 'Wolffian' sizes, I used a 'minus two stops' rule. Now with the M8, this is the only way to go.

 

Why? Because it is not the negative, or of course the image that falls on the sensor, that has to be sharp. IT IS THE PRINT, OR THE PROJECTED IMAGE, OR THE SCREEN IMAGE, that have to be sharp (if sharpness is what we demand). And all these viewing images have to be enlarged 33% more with the smaller sensor format, for any final print etc. size you want to do.

 

Now the upshot of this is of course that you can forget all about depth of field for any focal lengths except wide wide-angle lenses. Even for 35mm and longer, depth of field is so thin that you should forget about it. Point focus on the main subject. In case of a landscape or other view, focus on an interesting object in what may be called 'the distant foreground', stop down to 5.6 or at most 8 for best definition at that distance and take your chances. With superwides, do not fall for the temptation to stop down beyond 8, or diffraction will start to do you in.

 

Remember that the position of the sharpness plane is our most important tool for directing the viewer's eye to what OUR eye found most interesting---our reason for ever taking the picture. It would take an unusual subject, or supreme compositional skill, to make something interesting out of a picture where the foreground looks like glued directly onto the background.

 

The old man from the Age of 10.5cm Standard Lenses

[nhabedi]

Advertisement (gone after registration)

The d.o.f. scale was wrong already before the M8. It goes back to the 1920's when it was assumed that the 35mm negative would be enlarged at most 3x, to give a print to compete with a contact print from a 6x9cm roll film camera. But some mischievous spoilsports, including the redoubtable Dr. Paul Wolff, soon started to do prints in 30, 40 or even 50cm sizes. And today a simple test will convince you that the computational basis of the Leica d.o.f. tables, which have not changed for eighty years, are insufficient even for a decent 10x15cm (4x6'') print.

 

I don't buy that. The circle of confusion not only depends on the print size but also on the distance of the viewer from the print. There are "classical" rules about how far away from a print you should be when looking at it and obviously the distance increases with the print size. Of course, if you have a 50cm size print and look at it with a loupe, you will see things you aren't supposed to see. That's the equivalent of pixel-peeping in the analog world.

[robsteve]

Lars is correct. If you really need something in focus, focus on it rather than rely on DOf. This applies even with wide angles.

 

A few years ago I made some test pictures with a 15mm CV lens and the M8. If you really wanted distant objects to be sharp, you needed to have the lens at infinity. In further tests with the 15mm CV, I tested shooting subjects at about 2m and f5.6 - f8. If you did not have the focus exact, the image was not as sharp as it should be.

 

I have since given up on the 15mm VC and went to a 16-21mm WATE because you can focus it with the rangefinder and it is also much sharper.

 

Robert

[tashley]

I'm sorry, we have covered this before at least a couple of times, but here we go again:

 

The d.o.f. scale was wrong already before the M8. It goes back to the 1920's when it was assumed that the 35mm negative would be enlarged at most 3x, to give a print to compete with a contact print from a 6x9cm roll film camera. But some mischievous spoilsports, including the redoubtable Dr. Paul Wolff, soon started to do prints in 30, 40 or even 50cm sizes. And today a simple test will convince you that the computational basis of the Leica d.o.f. tables, which have not changed for eighty years, are insufficient even for a decent 10x15cm (4x6'') print.

 

Therefore experienced photographers used the 'minus one stop' rule recommended above already decades ago. So did I. Enlarging to 'Wolffian' sizes, I used a 'minus two stops' rule. Now with the M8, this is the only way to go.

 

Why? Because it is not the negative, or of course the image that falls on the sensor, that has to be sharp. IT IS THE PRINT, OR THE PROJECTED IMAGE, OR THE SCREEN IMAGE, that have to be sharp (if sharpness is what we demand). And all these viewing images have to be enlarged 33% more with the smaller sensor format, for any final print etc. size you want to do.

 

Now the upshot of this is of course that you can forget all about depth of field for any focal lengths except wide wide-angle lenses. Even for 35mm and longer, depth of field is so thin that you should forget about it. Point focus on the main subject. In case of a landscape or other view, focus on an interesting object in what may be called 'the distant foreground', stop down to 5.6 or at most 8 for best definition at that distance and take your chances. With superwides, do not fall for the temptation to stop down beyond 8, or diffraction will start to do you in.

 

Remember that the position of the sharpness plane is our most important tool for directing the viewer's eye to what OUR eye found most interesting---our reason for ever taking the picture. It would take an unusual subject, or supreme compositional skill, to make something interesting out of a picture where the foreground looks like glued directly onto the background.

 

The old man from the Age of 10.5cm Standard Lenses

 

This is extremely close to what I do and I think very good advice. But there is no substitute for taking out a new lens, when you first buy it, and putting it on your M8. Choose a scene with detail at 1 metre, 10 metres and infinity and focus on each of those but with the others in frame and using a tripod and shutter release, take a shot at each aperture.

 

Boring, but if you study the results you'll not then keep getting taken by surprise and you will have DOF preview in your head!

 

There's also a great app for the iPhone called DOF calc. I use it when things are critical and I'm using a camera/lens combo that's new to me but I would say that if you want to print above A3 then go one stop tighter for each doubling of paper size. Ish.

 

Tim

 

 

ps and remember that pursuant to what others have said, there is only one plane relative to the sensor that is truly in focus. Everything else that falls within the DOF that is acceptable to you at your print size and viewing distance is merely acceptably sharp....

[tashley]

Of course, if you have a 50cm size print and look at it with a loupe, you will see things you aren't supposed to see. That's the equivalent of pixel-peeping in the analog world.

 

But haven't you noticed that at exhibitions there's been a massive trend to print large and invite the viewer in very close? Classically we're supposed to assume that the average viewing distance is about equal to the diagonal of the print but I so often see people viewing from much closer, in both public and commercial galleries.

[nhabedi]

But haven't you noticed that at exhibitions there's been a massive trend to print large and invite the viewer in very close? Classically we're supposed to assume that the average viewing distance is about equal to the diagonal of the print but I so often see people viewing from much closer, in both public and commercial galleries.

 

Yeah, I've noticed that as well and I admit I also do this from time to time. Still, I think that doesn't make the DOF scales suddenly "wrong". As others have said, the plane of exact sharpness is a plane in the mathematical sense and thus has depth zero. The DOF marks are about the area of acceptable sharpness and that's a matter of convention which involves assumptions about average viewing distance. So, if the scale is correct for a 20cm print, it is also correct for a 5m print, as the viewing distance will change accordingly.

 

For the original question this is not really relevant, though. The relevant point is that if the DOF scale was OK for you on analog Leicas, then on the M8 you'll have to add a stop as you're effectively enlarging a crop of the image you would have taken with an MP/M7 to the same print size. But if you already considered the DOF scale to be wrong for analog Leicas, you shouldn't trust it on digital Leicas either, or adjust it accordingly, depending on your print sizes and how you expect your prints to be viewed.

[Albert4321]

That would mean when putting a 35mm back on a medium format camera (such as a Hassy 501), you loose most of your depth of field...

 

in short answer, yes

[aulia]

Thank you all for the enthusiastic replies.

 

The first part of my question was answered, and I am comfortable with using 1 stop less to get acceptable sharpness for print sizes that I make.

 

As for the second part of my question, is the distance scale on the M lenses still accurate on the M8?

 

Thanks

[tashley]

Thank you all for the enthusiastic replies.

 

The first part of my question was answered, and I am comfortable with using 1 stop less to get acceptable sharpness for print sizes that I make.

 

As for the second part of my question, is the distance scale on the M lenses still accurate on the M8?

 

Thanks

 

yes

[aulia]

yes

 

Thank you

[lars_bergquist]

I don't buy that. The circle of confusion not only depends on the print size but also on the distance of the viewer from the print. There are "classical" rules about how far away from a print you should be when looking at it and obviously the distance increases with the print size. Of course, if you have a 50cm size print and look at it with a loupe, you will see things you aren't supposed to see. That's the equivalent of pixel-peeping in the analog world.

Sorry, but this is nonsense. The arguments you are thinking about are carryovers from those days in the 1930's when people got hysterical about the supposedly 'unnatural perspective' of long-focus lenses. The 1/30th of a mm circle of confusion was adopted on the premise that the small prints envisaged would be viewed at normal reading distance, i.e. c. 30cm, and that under these conditions, a printed detail would be accepted as sharp if the circle of confusion *in the print* did not exceed .1mm.

 

We tend to view prints or other flat images from a distance about equal to the diagonal of the rectangle. Observations in art galleries and museums tend to confirm that. We do that as long as that diagonal is not shorter than our eyes' close focusing limit. Obviously, we do not normally try to look at a 10x15cm street lab print from 18 cm away! Therefore, the critical print size is no longer 6x9cm, as it was in 1925. It is 18x24, or 8x10'', or A4. After that size, the picture diagonal is longer than the close focusing limit, and we do spontaneously increase the viewing distance.

 

People who insist in viewing poster-size prints with magnifiers are obviously not interested in photography (let alone in images). They want something to gripe about, so they can sound superior. On the other hand, people who think that a .33mm c.o.c. is a useful criterion for an A4 print are obviously disregarding the testimony of their own eyes.

 

The old man from the Age of 6x9cm Contacts

[pgk]

The d.o.f. scale was wrong already before the M8.

 

Not wishing to be pedantic (and most certainly not offensive) BUT....

 

The depth of field scale is not 'wrong', as pointed out it is merely based on a set of parameters which many may no longer wish to operate to. If we want to have any sort of depth of field scale on our lenses (legacy or new) they have to be based on some set of parameters and changing the 'classical' scales would be an incredibly good way of creating vast amounts of confusion (and no doubt equally numbers of arguments, etc.). So for better or worse we have depth of field scales based on parameters which once made sense but probably are far less relevant today.

 

Perhaps a simple way of thinking about the existing scales are that they can be used as an indicator and their use adjusted according to the end output requirement of the image file. So many might want to adopt the 'scale for the next wider stop' as a matter of course, whilst others might go 'two stops wider'. Ultra critical users can go back to basics and calculate the exact depth of field that they can use based on their own workflow if this is really to be desired (I can't ever imagine doing so)!

 

But in practical terms, we still do have a useful and relevant set of 'guides' which we can use together with experience and forethought to determine the most relevant focus setting for any particular image.

 

(from a man from the pre-pixel peeping practical age!)

[Alberti]

Lars is correct. If you really need something in focus, focus on it rather than rely on DOf. This applies even with wide angles.

- - I have since given up on the 15mm VC and went to a 16-21mm WATE because you can focus it with the rangefinder and it is also much sharper.

Robert

 

I came to the same conclusion as Robert: a non-RF coupled [CV] lens cannot sensibly be set to a HV distance, by hand.

Lars is correct, image-wise. As standard enlargements go.

 

But in my experience we now see a so much reduced plane of focus (and enlarging it on the screen to 100% shows it) that we might have to recalculate the DoF - or even just disregard it. This 'enlargement' is out of proportion really. But is is the new digital era. technically, any postproduction (printing) might revert to that digital 'plane'. So it is tangible. The game of the DoF has been spoilt again (referring to Lars' comments). This new enlargement (based on pixels/mm) is based on the sensor and the [RAW] processor, not it's size.

 

With my 25mmVC set to 3 m and F5.6 I saw i a specific picture that the plane of clear focus went through the legs of my subject - the head being slightly oof. Really counter-intuitive.

For that reason I have a savingsaccount now open for the RF coupled 24mm3.8.

 

Now most of you remember the Tri-X and the 21mm SA lenses: everything in focus. Was it the depth of the film emulsion that cretated that effect? Or was it the way of handling the film (low resolution plus a grain-based enlargement) that made it winning? I would expect a 1250 ASA setting on the M8 (reduced lines per mm) to give just that same DOF quality again.

 

If so, a tip is to increase the sensitivity with the wide angles. On several occasions I am now setting the sharpening OFF just to obtain a 'better looking' image with a more natural rendering of the sharpness (in portraits . .).

alberti