Home
SearchSearch MemberlistMemberlist RegisterRegister ProfileProfile Log in to check your private messagesLog in to check your private messages Log inLog in
Hyperfocal/zone focusing confusion
View previous topic :: View next topic  


PostPosted: Mon May 18, 2015 8:38 pm    Post subject: Hyperfocal/zone focusing confusion Reply with quote

I have managed to confuse myself wrt hyperfocal focusing...

When looking at the lens barrel and figuring out a hyperfocal area one seems to get a straightforward suggestion. For eexample: 1 50mm lens may suggest that at f8 the depth of field lies between 10 and 30 feet. However, I invariably find that when I then check through my viewfinder/focusing screen, the actual in-focus area is much shallower.

I clearly don't understand the technique properly?

I can think of these issues:
1: The viewfinder reports correctly as to where the EXACT focal plane lies, but the area of "acceptable focus" lies within the stated depth of field
2: I misunderstand completely how hyperfocal distances work
3: errrr. other.


Can anybody educate me? I sometimes just want to be able to set my lens at a nice and deep depth of field and shoot away. For example when doing street photography...

Thanks!

Rudolf


PostPosted: Tue May 19, 2015 6:48 am    Post subject: Reply with quote

What exactly is the hyperfocal distance at any given setting depends on your personal definition of "sharp enough". Only objects exactly at the hyperfocal distance will be in perfect focus, everything in front and back (to infinity) will be progressively more blurry and the definition of hyperfocal distance assumes that even at infinity sharpness will still be "good enough". The requirements for "good enough" are different for different materials (film or digital), intended purpose, viewing distance, personal preference, etc. The markings on lenses are just general suggestions, not hard facts and may even be different on lenses from different manufacturers (different definition of "acceptable focus"). I find that for use on digital the markings on all lenses are way too optimistic.


PostPosted: Tue May 19, 2015 7:23 am    Post subject: Reply with quote

It may be that the settings on your lens are too optimistic, but hyperfocal distance settings are a very well accepted value and have been for decades. In fact, I was paging through the pdf of the Pentax LX's owner's manual just the other day, and there is a section in the manual that is dedicated just to hyperfocal settings. And that's a camera manual, not a lens manual.

The hyperfocal distance scale on a lens indicates sharpness for a given range -- most often used to determine the sharpness at a minimum distance to infinity. In other words, everything beyond that minimum distance is in focus. Are you sure you know how to read the scale properly? Let's say you have a wide angle lens, say a 28mm, and you have it set to f/8, and you want to use the hyperfocal scale such that everything from a minimum distance to infinity will be in focus. You align the infinity mark on the distance scale to the 8 on the hyperfocal scale -- then you look to see the value at the other 8 opposite the center alignment mark. Let's say that value is 10 feet or about 3 meters. This means that everything from 10 feet to infinity will be in focus. Now I'm just guessing at these numbers, but they're probably pretty close.

Hyperfocal values are established based on something called the Circle of Confusion. This is a point on the negative or slide that can vary in size, but the smaller the better. The Circle of Confusion means that anything, when the circle is optimally small, that is the size of that circle or smaller will be considered to be an in-focus detail. If the detail is larger than that circle, then it is considered to be not in focus.

The Circle of Confusion varies in size, depending on how well focused a lens is. The more precise the focus, the smaller the Circle of Confusion. But every lens will have its own smallest achievable Circle of Confusion -- which represents the limit of the lens's resolving power. Now, the hyperfocal scale is probably not determined using the smallest possible Circle of Confusion, but rather a circle of an agreed upon size that is small enough to provide sharp images within the hyperfocal range.

I personally depend on the hyperfocal scale, especially when I'm shooting with a wide angle lens, and I do it so I can maximize the amount of sharp detail in an image. Now, it is true that if you focus on an object within the hyperfocal range, it will be a distance that doesn't line up with the hyperfocal scale, but it will be a distance that falls within the hyperfocal range. So you're not losing anything by using the hyperfocal scale instead of the point of focus you've established.

The only time I've had a problem with this scale was when I was trying out a pair of 17mm lenses on my NEX 7 (a Tamron and a Vivitar). With these lenses, the hyperfocal scales were of no use at all, but then neither were the focusing scales, either. Those lenses just don't work on a crop body sensor. (I tried the Tamron on my crop body EOS too and it was just as bad as the NEX). So there are exceptions, but they're special cases and not the norm by any means.


PostPosted: Tue May 19, 2015 12:02 pm    Post subject: Reply with quote

miran wrote:
I find that for use on digital the markings on all lenses are way too optimistic.


You are absolutely right!

Yesterday I made some pictures (NEX & VC 15mm) and used the "focal scale" on the lens in combination with the aperture used. I found that many of the pictures have been out of focus in rather important distances within the pictures if watched at 100% view. However, for small prints up to 10x15cms (typical holiday pictures) still good enough.

Lesson learned. Never trust that scale! Not even on super wides.


PostPosted: Tue May 19, 2015 12:47 pm    Post subject: Reply with quote

Just to show you the difference of two pictures both well within the limits on the scale (as previously mentioned). Both taken with VC 15mm & NEX. 100% crops from the center of the images. 1st one is rather perfect and 2nd one just crap.






PostPosted: Tue May 19, 2015 12:51 pm    Post subject: Reply with quote

Keep in mind also that those markings were made for 35mm frame (so, when you take a picture on aps-c and watch it somewhere, you are basically enlarging it more than it was assumed when calculations were made), and calculating a "standard" print size (I knew that size, but I don't remember), which probably was different from a 29" dispaly seen from 40 cm away (that's my setup when I pixel peep, more or less).
So you have to be conservative with those markings: usually I use the ones from a stop wider aperture, and it kind of works.


PostPosted: Tue May 19, 2015 12:54 pm    Post subject: Reply with quote

cooltouch wrote:
It may be that the settings on your lens are too optimistic, but hyperfocal distance settings are a very well accepted value and have been for decades. In fact, I was paging through the pdf of the Pentax LX's owner's manual just the other day, and there is a section in the manual that is dedicated just to hyperfocal settings. And that's a camera manual, not a lens manual.

The hyperfocal distance scale on a lens indicates sharpness for a given range -- most often used to determine the sharpness at a minimum distance to infinity. In other words, everything beyond that minimum distance is in focus. Are you sure you know how to read the scale properly? Let's say you have a wide angle lens, say a 28mm, and you have it set to f/8, and you want to use the hyperfocal scale such that everything from a minimum distance to infinity will be in focus. You align the infinity mark on the distance scale to the 8 on the hyperfocal scale -- then you look to see the value at the other 8 opposite the center alignment mark. Let's say that value is 10 feet or about 3 meters. This means that everything from 10 feet to infinity will be in focus. Now I'm just guessing at these numbers, but they're probably pretty close.



I am reading the scale correctly, but just finding a very clear falloff in sharpness ,away from the focal plane Yet WITHIN the hyperfocal distance.

I wonder if it was just not so noticeable on film?


PostPosted: Tue May 19, 2015 12:55 pm    Post subject: Reply with quote

Aanything wrote:
Keep in mind also that those markings were made for 35mm frame (so, when you take a picture on aps-c and watch it somewhere, you are basically enlarging it more than it was assumed when calculations were made), and calculating a "standard" print size (I knew that size, but I don't remember), which probably was different from a 29" dispaly seen from 40 cm away (that's my setup when I pixel peep, more or less).
So you have to be conservative with those markings: usually I use the ones from a stop wider aperture, and it kind of works.


Yeah that rings true to my ears.
Smile


PostPosted: Tue May 19, 2015 1:06 pm    Post subject: Reply with quote

rudolfkremers wrote:
Aanything wrote:
Keep in mind also that those markings were made for 35mm frame (so, when you take a picture on aps-c and watch it somewhere, you are basically enlarging it more than it was assumed when calculations were made), and calculating a "standard" print size (I knew that size, but I don't remember), which probably was different from a 29" dispaly seen from 40 cm away (that's my setup when I pixel peep, more or less).
So you have to be conservative with those markings: usually I use the ones from a stop wider aperture, and it kind of works.


Yeah that rings true to my ears.
Smile


Did you ever watch your pictures from film in a size of approx. 200x140cms? That would approximately be the equivalent to the 100% crop on a Full HD screen 28 inches. Nontheless, it was more critical on 24x36 format as the DOF is more shallow than on DX-format.

However, I'll rather trust the enlarged view for focusing than the scale on the lens. That's my conclusion.


PostPosted: Tue May 19, 2015 1:30 pm    Post subject: Reply with quote

tb_a wrote:


Did you ever watch your pictures from film in a size of approx. 200x140cms? That would approximately be the equivalent to the 100% crop on a Full HD screen 28 inches. Nontheless, it was more critical on 24x36 format as the DOF is more shallow than on DX-format.

However, I'll rather trust the enlarged view for focusing than the scale on the lens. That's my conclusion.


DOF on FF is smaller than on crop if the field of view is the same, because you are at a closer distance from your subject.
At the same distance from the subject (that's just how I understood it: I see the logic of this, but I ignore most of the math behind this. It works empirically for me, though, so I take it for good), and with prints of the same size, dof with a crop format would indeed be smaller, as the image is enlarged more, so the "acceptably in focus area" is thinner.
I watched pictures from film of approx 200x140 cms, but i suspect they weren't taken relying upon the hyperfocus markings on the lens (and now that I think about it, probably they weren't taken in 35mm either): this is another case in which they (the marks) would appear too optimistic.


PostPosted: Tue May 19, 2015 1:33 pm    Post subject: Reply with quote

tb_a wrote:

However, I'll rather trust the enlarged view for focusing than the scale on the lens. That's my conclusion.


That's exactly what I do too, when I want to be sure of the dof limits.
But sometimes is fun to just slap a 18mm on the camera, set the aperture to f11, and just shoot from the hip.


PostPosted: Tue May 19, 2015 1:45 pm    Post subject: Reply with quote

Aanything wrote:

But sometimes is fun to just slap a 18mm on the camera, set the aperture to f11, and just shoot from the hip.


And that is exactly what I've done yesterday with the CV 15mm on my NEX and you have already seen the "not so good" results at least in "crop-view" mode. Wink
Maybe I should have taken the CV 12mm lens instead. Unfortunately the 12mm lens is not as tiny as the 15mm one. But I will repeat the test with 12mm. At least in theory it must work better.


PostPosted: Thu May 21, 2015 7:38 am    Post subject: Reply with quote

Hi Rudolf,

Hyperfocal distance is not "an area". As said above, it is a calculated point, dependent on focal length and aperture setting. The definition is "the distance between the camera and the nearest point that is still acceptably sharp when the lens is focused on infinity". It should not be confused with DoF (although that is part of it).

Taking your example, a 50mm lens, set at f8 would give a hyperfocal distance of 12.5 metres. So, if ye set the lens at 12.5 metres, then everything from half that distance(6.25 metres) to infinity will have an acceptably sharp DoF. Re. FF and APS-C, a 50mm lens is still a 50mm lens, the equivalent angle of view changes (narrows by the "crop factor" for each lens) not the focal length.

There are formulae tae use where ye can calculate DoF for all yer lenses individually. Or download Hyperfocal distance tables.

Formula for Hyperfocal distance - H = F/cf

H=Hyperfocal distance
F=focal length
C=Circle of confusion
F=f number

Formula for DoF - Dn = HD/H+D - Df = HD/H-D

Dn+Df are near and far limits of DoF
H=Hyperfocal distance
D=Focused distance

So, DoF for a 50mm lens @f8, focused at 5 metres is...

12.5 x 5 divided by 12.5+5 = 3.57 metres
12.5 x 5 divided by 12.5-5 = 8.33 metres

DoF extends from 3.57 metres tae 8.33 metres. Cool


PostPosted: Thu May 21, 2015 9:54 am    Post subject: Reply with quote

TAo2 wrote:

Taking your example, a 50mm lens, set at f8 would give a hyperfocal distance of 12.5 metres. So, if ye set the lens at 12.5 metres, then everything from half that distance(6.25 metres) to infinity will have an acceptably sharp DoF. Re. FF and APS-C, a 50mm lens is still a 50mm lens, the equivalent angle of view changes (narrows by the "crop factor" for each lens) not the focal length.


I think the theory is most of the people clear anyway. At least for me it's more than clear, having read and learned this all several decades ago, when photography was purely manual with no automated features available what so ever.
However, theory and daily practice are two different issues and to talk about the differences in more understandable words is also something different.

It goes without saying that 50mm is always 50mm irrespective of the sensor size. However, as the FOV is quite different you have to take the picture from different distances, to reach the same FOV. This leads automatically to larger DOF on smaller sensor sizes AT SAME FOV.
That is the logic behind, when I state that the DOF is bigger on DX sensor compared to FF sensor. Just to make clear, that I know what's the issue.

Nontheless is the only aid which is available on the street either the scale on the lens or the viewer of the camera. The rest depends on your feeling and personal experience based on the systematic of the physical formula behind it and not the formula itself.

That is at least my point of view.


PostPosted: Thu May 21, 2015 10:02 am    Post subject: Reply with quote

Ok so even though I am potentially more confused rather than less Very Happy I still end up with the same practical question.

The lens marking tells me that the acceptably sharp area is as wide as you describe, yet in practice I fond that when I actually test this there is a clear fall off of sharpness even within the range that is supposedly acceptably sharp.

Now, I fully understand that the only way to really test that is to actually TEST that with real photography. Smile
I was just wondering if I understood the practical application. Which I think I do.

I have seen so many comments on famous photos where the photographer set his/her camera to focus between a certain distance and infinity and just trust that is in focus, that I wanted to make use of that myself. Rather useful in situations where you can't quickly focus! Also pretty important in landscape photography or any other photography where I want to get as much of a scene in focus as I can.

R


PostPosted: Thu May 21, 2015 12:13 pm    Post subject: Reply with quote

Rudolf,

As I've proved some postings above, the difference of quality of presented pictures mainly depends on the size of presentation.
The larger the intended size for presentation will be, the less the "average" method is recommended.
In the analog film times we have all been less demanding, unless a real big slide presentation was intended. For normal paper pictures any difference of super sharp compared to "within limits" was barely visible.
The more people intend to crop their pictures for larger view of the more important parts of their pictures, the more the problem will increase. That was hardly done in film times, that's also a reason for the different perception of the same issue.
That's more or less the whole story about it.


PostPosted: Thu May 21, 2015 2:10 pm    Post subject: Reply with quote

tb_a wrote:
Rudolf,

As I've proved some postings above, the difference of quality of presented pictures mainly depends on the size of presentation.
The larger the intended size for presentation will be, the less the "average" method is recommended.
In the analog film times we have all been less demanding, unless a real big slide presentation was intended. For normal paper pictures any difference of super sharp compared to "within limits" was barely visible.
The more people intend to crop their pictures for larger view of the more important parts of their pictures, the more the problem will increase. That was hardly done in film times, that's also a reason for the different perception of the same issue.
That's more or less the whole story about it.


Yes I get that now Smile

My take away is as follows:
My original understanding was flawed, as I assumed the in focus area would be equally in focus across the dof.
I can still use this technique, especially on film, but be somewhat conservative in expectations and expect some sharpness fall off away from the focal plane.

Very helpful thread!

Thanks all!


PostPosted: Thu May 21, 2015 6:29 pm    Post subject: Reply with quote

TAo2 wrote:
Hi Rudolf,

Hyperfocal distance is not "an area". As said above, it is a calculated point, dependent on focal length and aperture setting. The definition is "the distance between the camera and the nearest point that is still acceptably sharp when the lens is focused on infinity". It should not be confused with DoF (although that is part of it).

Taking your example, a 50mm lens, set at f8 would give a hyperfocal distance of 12.5 metres. So, if ye set the lens at 12.5 metres, then everything from half that distance(6.25 metres) to infinity will have an acceptably sharp DoF. Re. FF and APS-C, a 50mm lens is still a 50mm lens, the equivalent angle of view changes (narrows by the "crop factor" for each lens) not the focal length.

There are formulae tae use where ye can calculate DoF for all yer lenses individually. Or download Hyperfocal distance tables.

Formula for Hyperfocal distance - H = F/cf

H=Hyperfocal distance
F=focal length
C=Circle of confusion
F=f number

Formula for DoF - Dn = HD/H+D - Df = HD/H-D

Dn+Df are near and far limits of DoF
H=Hyperfocal distance
D=Focused distance

So, DoF for a 50mm lens @f8, focused at 5 metres is...

12.5 x 5 divided by 12.5+5 = 3.57 metres
12.5 x 5 divided by 12.5-5 = 8.33 metres

DoF extends from 3.57 metres tae 8.33 metres. Cool


Your formulas are confusing as written. It took me a while to figure out that your "Formula for DoF" is actually two formulae and you used a minus sign as a separator. So, it would be clearer to write them as:

Dn = HD / (H + D)
Df = HD / (H - D)

Also, you have F written twice, for two different variable values. Although it appears that you used it only once, F = Focal length, in this case. Let's let N = f number. If you would like to calculate DoF by itself, this formula can be used:

DoF = (2NC[F^2][D^2])/([F^4]-[N^2][C^2][D^2)]

Here it is, lifted from Wikipedia's page on the subject,, using their variable values. It's more legible to read this way:


Note that Wikipedia also shows a wavy equals sign in their formula. It means the value derived from the formula is an approximate one. I believe this is because the Circle of Confusion can be subject to some interpretation.

This entire subject is quite interesting and can get quite involved. Wikipedia goes into considerable depth on the topics of Hyperfocal Distance and Depth of Field. See:


http://en.wikipedia.org/wiki/Hyperfocal_distance
http://en.wikipedia.org/wiki/Depth_of_field


PostPosted: Thu May 21, 2015 7:29 pm    Post subject: Reply with quote

Hi CT,

Where do ah begin? Ye really think that.....

DoF = (2NC[F^2][D^2])/([F^4]-[N^2][C^2][D^2)] is easier tae understand than...


Formula for DoF - Dn = HD/H+D - Df = HD/H-D

Dn+Df are near and far limits of DoF
H=Hyperfocal distance
D=Focused distance
??


and that ...



is more legible? More legible for a mathematician, perhaps...

The minus signs are not minus signs - they are dashes - punctuation marks.

The DoF formula is one formula, consisting of 2 logical parts - near and far limits of a particular DoF, not 2 formulae

The second capital F should have been lower case, it was an unnoticed typo and for that - I deeply apologise... Cool


PostPosted: Fri May 22, 2015 2:13 am    Post subject: Reply with quote

As a matter of fact, I do think what I wrote out is easier to follow. You just need to parse the brackets, which I put in there for clarity, after all. It would have been much clearer if my keyboard would respond the way I write -- which is why I cut and pasted the formula from Wikipedia -- to make it clearer.

When you write out a formula, I would suggest you think about using different punctuation marks than dashes -- because of their double meaning. Commas, semicolons, elipses -- any of these would have been better than minus signs -- ahem, dashes. I spent a good 15 or 20 minutes trying to figure out what was going on with your formula until I realized they were punctuation! More legible indeed!

If you want to call it one formula, fine, but it's two equations, then, because it's two separate calculations.


PostPosted: Sun May 24, 2015 10:04 am    Post subject: Reply with quote

I have found this site which may be very helpful for further understanding of the topic:

http://www.dofmaster.com/dofjs.html

Unfortunately my Ricoh GXR-M isn't listed there... Sad
However, it's a nice feature anyway.

Have fun.