Need help with focusing issue.

Started 3 months ago | Questions
Nas from the Valley Regular Member • Posts: 117
Re: Need help with focusing issue.
1

charley5 wrote:

Hi Everyone:

I am having a frustrating time with a focusing issue and I need help to solve it. I focus on the face of the person initially by pressing down the shutter half way. Then, because I want the face to be at the top third of the frame, I move the camera (with the shutter still half-way depressed) usually to chest level, and press all the way down. Their chest then is well focused in the final image but their face is not. This is really frustrating. I have lost several good shots in this way. What may be the problem? I use single point focusing and 9 squares. Thanks for any suggestions. Here are two examples:

I couldn't read all of the comments above but of course you will have a problem with focusing. It is obvious!

When you focus (in single focus mode) and hold the shutter button and move the camera you are changing focus plane! Don't expect camera to remember where exactly you focused and lock that focus spot forever!

Let me give you an example: you focus on the face which is exactly let's say 2.37 meters away from the camera and focus plane will stay on the same distance (2.37 meters) as long as you are holding the shutter button. Then you move down the camera a little bit and the face will be on let's say 2.32 meters away from the camera but the chest now will be on 2.37 meters. So the focus plane will be changed and the face will no longer be the sharpest spot since it's not on the focus plane anymore. Simple as that.

The only way to do it the way you want it is with wide tracking, face/eye detection or the best is just use the focus points situated at the top of the sensor where the face will be.

Focal plane changes when you re-focus.

I know drawing is funny but that's what I wanted to show you. The red line is the focal plane.

baobob
baobob Forum Pro • Posts: 13,994
Re: Need help with focusing issue.

Not the best way for portrait, the best being  a small one on the eye controlling for DOF

depending on FL and distance DOF can be very very short with eye sharp and nose or ear OOF

-- hide signature --

Good judgment comes from experience
Experience comes from bad judgment

 baobob's gear list:baobob's gear list
Sony RX100 Olympus Tough TG-4 Panasonic ZS200 Fujifilm X-H1 Fujifilm X-T3 +10 more
Nas from the Valley Regular Member • Posts: 117
Re: Need help with focusing issue.

baobob wrote:

Not the best way for portrait, the best being a small one on the eye controlling for DOF

depending on FL and distance DOF can be very very short with eye sharp and nose or ear OOF

The problem is not the zone focusing he used, it will be the same if he uses even the smallest focusing point. The problem is that he moves the camera down and focus plane changes.

Erik Baumgartner Senior Member • Posts: 1,428
Re: Need help with focusing issue.

charley5 wrote:

Bob Tullis wrote:

charley5 wrote:

Hi Everyone:

I am having a frustrating time with a focusing issue and I need help to solve it. I focus on the face of the person initially by pressing down the shutter half way. Then, because I want the face to be at the top third of the frame, I move the camera (with the shutter still half-way depressed) usually to chest level, and press all the way down. Their chest then is well focused in the final image but their face is not. This is really frustrating. I have lost several good shots in this way. What may be the problem? I use single point focusing and 9 squares. Thanks for any suggestions. Here are two examples:

Not really a focusing issue, it's an operator thing. . . can you not use a single AF point and move the AF box to the face?

I don't see a lot wrong with these, anyway.

Hi. The faces are soft and the chest area is sharp. I will place the focus area on the face, it is just that it is time consuming and kids are not patient!

You can move the focus box to where you want the face to be in the frame before you even raise the camera. If I’m shooting people I will move the the focus box to the top third of the frame for both vertical and horizontal orientations - the camera can be set to remember where you set the point for each orientation so that you don’t have keep moving it every time you flip your camera around. It can speed things up a bit - your focus box will always be in the ballpark to start.

 Erik Baumgartner's gear list:Erik Baumgartner's gear list
Fujifilm X-T2 Fujifilm X-T20 Fujifilm XF 35mm F1.4 R Fujifilm XF 56mm F1.2 R Fujifilm XF 50-140mm F2.8 +1 more
Jerry-astro
MOD Jerry-astro Forum Pro • Posts: 14,785
Re: Need help with focusing issue.

Nas from the Valley wrote:

charley5 wrote:

Hi Everyone:

I am having a frustrating time with a focusing issue and I need help to solve it. I focus on the face of the person initially by pressing down the shutter half way. Then, because I want the face to be at the top third of the frame, I move the camera (with the shutter still half-way depressed) usually to chest level, and press all the way down. Their chest then is well focused in the final image but their face is not. This is really frustrating. I have lost several good shots in this way. What may be the problem? I use single point focusing and 9 squares. Thanks for any suggestions. Here are two examples:

I couldn't read all of the comments above but of course you will have a problem with focusing. It is obvious!

When you focus (in single focus mode) and hold the shutter button and move the camera you are changing focus plane! Don't expect camera to remember where exactly you focused and lock that focus spot forever!

Let me give you an example: you focus on the face which is exactly let's say 2.37 meters away from the camera and focus plane will stay on the same distance (2.37 meters) as long as you are holding the shutter button. Then you move down the camera a little bit and the face will be on let's say 2.32 meters away from the camera but the chest now will be on 2.37 meters. So the focus plane will be changed and the face will no longer be the sharpest spot since it's not on the focus plane anymore. Simple as that.

The only way to do it the way you want it is with wide tracking, face/eye detection or the best is just use the focus points situated at the top of the sensor where the face will be.

Focal plane changes when you re-focus.

I know drawing is funny but that's what I wanted to show you. The red line is the focal plane.

The point you're making is spot on, however, it's worth noting again that the focal plane is actually curved, not straight as you've illustrated, and equidistant to the camera.

-- hide signature --

Jerry-Astro
Fujifilm X Forum Co-Mod

 Jerry-astro's gear list:Jerry-astro's gear list
Fujifilm X-H1 Carl Zeiss Touit 2.8/12 Fujifilm XF 10-24mm F4 R OIS Fujifilm 16-55mm F2.8R LM WR Fujifilm XF 100-400mm F4.5-5.6 OIS WR +1 more
Truman Prevatt
Truman Prevatt Veteran Member • Posts: 9,939
Re: Need help with focusing issue.

FujiJon wrote:

Are you using continuous focus? If so, switch to single focus. Also, make sure the camera is set to "focus" not "release". It may help to select the focus point over the subject's face, rather than focus-recompose, or use face/eye detection (the latter is easier and quicker if you are photographing people)

Good advice.  The OP wants to insure AF-S and focus not release.  The best way would be to put the focus point over the subjects face instead of focus recompose.

However, I'm not sure the focus-recompose is the issue here.  I seriously doubt the lens is a flat field lens.  Secondly at the focal length and f-stop it would not matter anyway.  In realty I'm not sure that overall the entire image (especially the first on) is not soft throughout. The second one - one of the buttons seems to be in focus, however, even there it is a little soft.

Given this lens did not fair too well in some of the previous post - I would wonder about the lens.

-- hide signature --

Truman
www.pbase.com/tprevatt

 Truman Prevatt's gear list:Truman Prevatt's gear list
Fujifilm X-Pro2 Fujifilm X-H1 Fujifilm XF 18mm F2 R Fujifilm XF 35mm F1.4 R Fujifilm XF 55-200mm F3.5-4.8 R LM OIS +9 more
ttusteveo Regular Member • Posts: 262
Adequate depth of field for Focus/Recompose

Jerry-astro wrote:

Nas from the Valley wrote:

charley5 wrote:

Hi Everyone:

I am having a frustrating time with a focusing issue and I need help to solve it. I focus on the face of the person initially by pressing down the shutter half way. Then, because I want the face to be at the top third of the frame, I move the camera (with the shutter still half-way depressed) usually to chest level, and press all the way down. Their chest then is well focused in the final image but their face is not. This is really frustrating. I have lost several good shots in this way. What may be the problem? I use single point focusing and 9 squares. Thanks for any suggestions. Here are two examples:

I couldn't read all of the comments above but of course you will have a problem with focusing. It is obvious!

When you focus (in single focus mode) and hold the shutter button and move the camera you are changing focus plane! Don't expect camera to remember where exactly you focused and lock that focus spot forever!

Let me give you an example: you focus on the face which is exactly let's say 2.37 meters away from the camera and focus plane will stay on the same distance (2.37 meters) as long as you are holding the shutter button. Then you move down the camera a little bit and the face will be on let's say 2.32 meters away from the camera but the chest now will be on 2.37 meters. So the focus plane will be changed and the face will no longer be the sharpest spot since it's not on the focus plane anymore. Simple as that.

The only way to do it the way you want it is with wide tracking, face/eye detection or the best is just use the focus points situated at the top of the sensor where the face will be.

Focal plane changes when you re-focus.

I know drawing is funny but that's what I wanted to show you. The red line is the focal plane.

The point you're making is spot on, however, it's worth noting again that the focal plane is actually curved, not straight as you've illustrated, and equidistant to the camera.

Question...

An online depth-of-field calculator shows that for the first image:
Focal length 39mm, f/8, that the depth of field

I think he had to be at least 6 feet from his subject when he took the photo, but the OP can confirm.

At a distance of 8 feet, the depth of field would have been:
In front of subject: 1.58 ft
Behind subject: 2.61 ft

Wouldn't that have been sufficient for Focus+Recompose to have captured sharp detail, in spite of the slight change in the focal plane?

 ttusteveo's gear list:ttusteveo's gear list
Fujifilm X-T2 Fujifilm XF 14mm F2.8 R Fujifilm XF 18-55mm F2.8-4 R LM OIS Samyang 12mm F2.0 NCS CS Fujifilm XF 80mm F2.8 Macro
Truman Prevatt
Truman Prevatt Veteran Member • Posts: 9,939
Re: Can’t say I agree...

Jerry-astro wrote:

Simon Elwell wrote:

This is typical! You use the focus and recompose method - a tried and trusted approach for many photographers. All but one of the responses so far have told you to change your technique - and use *their* approach - also tried and trusted.

But why should you have to?

You don't. Focus and recompose is not only a perfectly acceptable way of focusing and framing - but it works just fine in Fuji cameras - as one poster pointed out - you have to get the settings right.

Except for one little problem... that’s not necessarily true. Portraits are often taken at large apertures with very thin DOF in order to isolate the subject. At those apertures, even a slight focus miss can throw your image OOF. The camera focuses at a fixed range from the camera/lens, which means that the precise focus point is not linear, it’s actually circular — an arc at a fixed distance from the camera. With very thin DOF, moving the camera even a little after establishing focus can throw the location you’ve moved the camera to outside the zone that is in focus. Granted, this may be unlikely unless you’re shooting at very large apertures, but it can definitely happen... and it’s happened to me in the past.

This all true for highest quality flat field lenses. I would not place the 16-80 in that category. In reality there are very few lenses that are truly flat field. It is difficult to design a lens and eliminate field curvature. Focus recompose actually is the best way to focus on a spherical surface.

A single lens will focus all objects from the same distance - independent on the incidence angle. That is the focal field will be the surface of a sphere and not a flat field (plane). To create a flat field lens would require a lens with different focal lengths for different angles of incidence on the lens. That requires a lots of expensive glass to do so. The only true flat field lenses are copy lenses and enlarger lenses. Macro lenses are next. However, macro lenses are big, heavy and slow - all a result of the flat field design requirements. Wide angle lenses suffer more from curvature than longer lenses which is why most macro lenses are telephoto.

However, even the Fuji X80 f2.8 as big, heavy, slow and expensive as it is has some detectable field curvature.

http://www.sansmirror.com/lenses/lens-reviews/lenses-for-fujifilm/fujifilm-80mm-f28-lens.html

Eliminating field curvature is a a tough nut to crack. The real solution is a curved sensor.  Sony has patented a medium format curved sensor and Nikon has a FF curved sensor. That would really simplify the lens design and result in smaller and less expensive lenses.

Corner softness wide open is the prime indicator of a curved focal field. If one has to stop a lens down to 5.6 or 8 to increase corner sharpness, the field is not flat.

While if one has the time, it might be better to place the focus box over the face in a portrait. However, I don't expect the 16-80 has a sufficiently flat field to eliminate focus recompose especially at f5.8 and f8 which is where these images were taken at a pretty "normal" focal length.

I don't think focus recompose was an issue with the OP's images. The images for the most part soft all over.

Optical field curvature.

https://www.discoverdigitalphotography.com/2015/lens-problems-field-curvature/

https://photographylife.com/what-is-field-curvature

http://emlab.utep.edu/IntroOpticsPhotonics/Summary%20of%20Field%20Curvature.pdf

Examples of field curvature.

https://petapixel.com/2016/12/21/field-curvature-tricky-problem-photography/

-- hide signature --

Truman
www.pbase.com/tprevatt

 Truman Prevatt's gear list:Truman Prevatt's gear list
Fujifilm X-Pro2 Fujifilm X-H1 Fujifilm XF 18mm F2 R Fujifilm XF 35mm F1.4 R Fujifilm XF 55-200mm F3.5-4.8 R LM OIS +9 more
Jerry-astro
MOD Jerry-astro Forum Pro • Posts: 14,785
Re: Adequate depth of field for Focus/Recompose

ttusteveo wrote:

Jerry-astro wrote:

Nas from the Valley wrote:

charley5 wrote:

Hi Everyone:

I am having a frustrating time with a focusing issue and I need help to solve it. I focus on the face of the person initially by pressing down the shutter half way. Then, because I want the face to be at the top third of the frame, I move the camera (with the shutter still half-way depressed) usually to chest level, and press all the way down. Their chest then is well focused in the final image but their face is not. This is really frustrating. I have lost several good shots in this way. What may be the problem? I use single point focusing and 9 squares. Thanks for any suggestions. Here are two examples:

I couldn't read all of the comments above but of course you will have a problem with focusing. It is obvious!

When you focus (in single focus mode) and hold the shutter button and move the camera you are changing focus plane! Don't expect camera to remember where exactly you focused and lock that focus spot forever!

Let me give you an example: you focus on the face which is exactly let's say 2.37 meters away from the camera and focus plane will stay on the same distance (2.37 meters) as long as you are holding the shutter button. Then you move down the camera a little bit and the face will be on let's say 2.32 meters away from the camera but the chest now will be on 2.37 meters. So the focus plane will be changed and the face will no longer be the sharpest spot since it's not on the focus plane anymore. Simple as that.

The only way to do it the way you want it is with wide tracking, face/eye detection or the best is just use the focus points situated at the top of the sensor where the face will be.

Focal plane changes when you re-focus.

I know drawing is funny but that's what I wanted to show you. The red line is the focal plane.

The point you're making is spot on, however, it's worth noting again that the focal plane is actually curved, not straight as you've illustrated, and equidistant to the camera.

Question...

An online depth-of-field calculator shows that for the first image:
Focal length 39mm, f/8, that the depth of field

I think he had to be at least 6 feet from his subject when he took the photo, but the OP can confirm.

At a distance of 8 feet, the depth of field would have been:
In front of subject: 1.58 ft
Behind subject: 2.61 ft

Wouldn't that have been sufficient for Focus+Recompose to have captured sharp detail, in spite of the slight change in the focal plane?

If you were responding to my comments, they were intended to be more generic as opposed to relating to this specific example.  The larger the aperture and thinner the DOF, the more likely you are to run into issues with focus and recompose.  Shooting portraits, for example, with the 56/1.2 wide open, can most definitely get one in trouble with focus and recompose as opposed to simply moving the focus point directly on your intended point of focus.

-- hide signature --

Jerry-Astro
Fujifilm X Forum Co-Mod

 Jerry-astro's gear list:Jerry-astro's gear list
Fujifilm X-H1 Carl Zeiss Touit 2.8/12 Fujifilm XF 10-24mm F4 R OIS Fujifilm 16-55mm F2.8R LM WR Fujifilm XF 100-400mm F4.5-5.6 OIS WR +1 more
Jerry-astro
MOD Jerry-astro Forum Pro • Posts: 14,785
Re: Can’t say I agree...

Truman Prevatt wrote:

Jerry-astro wrote:

Simon Elwell wrote:

This is typical! You use the focus and recompose method - a tried and trusted approach for many photographers. All but one of the responses so far have told you to change your technique - and use *their* approach - also tried and trusted.

But why should you have to?

You don't. Focus and recompose is not only a perfectly acceptable way of focusing and framing - but it works just fine in Fuji cameras - as one poster pointed out - you have to get the settings right.

Except for one little problem... that’s not necessarily true. Portraits are often taken at large apertures with very thin DOF in order to isolate the subject. At those apertures, even a slight focus miss can throw your image OOF. The camera focuses at a fixed range from the camera/lens, which means that the precise focus point is not linear, it’s actually circular — an arc at a fixed distance from the camera. With very thin DOF, moving the camera even a little after establishing focus can throw the location you’ve moved the camera to outside the zone that is in focus. Granted, this may be unlikely unless you’re shooting at very large apertures, but it can definitely happen... and it’s happened to me in the past.

This all true for highest quality flat field lenses. I would not place the 16-80 in that category. In reality there are very few lenses that are truly flat field. It is difficult to design a lens and eliminate field curvature. Focus recompose actually is the best way to focus on a spherical surface.

A single lens will focus all objects from the same distance - independent on the incidence angle. That is the focal field will be the surface of a sphere and not a flat field (plane). To create a flat field lens would require a lens with different focal lengths for different angles of incidence on the lens. That requires a lots of expensive glass to do so. The only true flat field lenses are copy lenses and enlarger lenses. Macro lenses are next. However, macro lenses are big, heavy and slow - all a result of the flat field design requirements. Wide angle lenses suffer more from curvature than longer lenses which is why most macro lenses are telephoto.

However, even the Fuji X80 f2.8 as big, heavy, slow and expensive as it is has some detectable field curvature.

http://www.sansmirror.com/lenses/lens-reviews/lenses-for-fujifilm/fujifilm-80mm-f28-lens.html

Eliminating field curvature is a a tough nut to crack. The real solution is a curved sensor. Sony has patented a medium format curved sensor and Nikon has a FF curved sensor. That would really simplify the lens design and result in smaller and less expensive lenses.

Corner softness wide open is the prime indicator of a curved focal field. If one has to stop a lens down to 5.6 or 8 to increase corner sharpness, the field is not flat.

While if one has the time, it might be better to place the focus box over the face in a portrait. However, I don't expect the 16-80 has a sufficiently flat field to eliminate focus recompose especially at f5.8 and f8 which is where these images were taken at a pretty "normal" focal length.

I don't think focus recompose was an issue with the OP's images. The images for the most part soft all over.

Optical field curvature.

https://www.discoverdigitalphotography.com/2015/lens-problems-field-curvature/

https://photographylife.com/what-is-field-curvature

http://emlab.utep.edu/IntroOpticsPhotonics/Summary%20of%20Field%20Curvature.pdf

Examples of field curvature.

https://petapixel.com/2016/12/21/field-curvature-tricky-problem-photography/

My comments were more of a generic nature, Truman, not specifically intended to be in reference to the OPs examples.

-- hide signature --

Jerry-Astro
Fujifilm X Forum Co-Mod

 Jerry-astro's gear list:Jerry-astro's gear list
Fujifilm X-H1 Carl Zeiss Touit 2.8/12 Fujifilm XF 10-24mm F4 R OIS Fujifilm 16-55mm F2.8R LM WR Fujifilm XF 100-400mm F4.5-5.6 OIS WR +1 more
Truman Prevatt
Truman Prevatt Veteran Member • Posts: 9,939
Re: Adequate depth of field for Focus/Recompose

ttusteveo wrote:

Jerry-astro wrote:

Nas from the Valley wrote:

charley5 wrote:

Hi Everyone:

I am having a frustrating time with a focusing issue and I need help to solve it. I focus on the face of the person initially by pressing down the shutter half way. Then, because I want the face to be at the top third of the frame, I move the camera (with the shutter still half-way depressed) usually to chest level, and press all the way down. Their chest then is well focused in the final image but their face is not. This is really frustrating. I have lost several good shots in this way. What may be the problem? I use single point focusing and 9 squares. Thanks for any suggestions. Here are two examples:

I couldn't read all of the comments above but of course you will have a problem with focusing. It is obvious!

When you focus (in single focus mode) and hold the shutter button and move the camera you are changing focus plane! Don't expect camera to remember where exactly you focused and lock that focus spot forever!

Let me give you an example: you focus on the face which is exactly let's say 2.37 meters away from the camera and focus plane will stay on the same distance (2.37 meters) as long as you are holding the shutter button. Then you move down the camera a little bit and the face will be on let's say 2.32 meters away from the camera but the chest now will be on 2.37 meters. So the focus plane will be changed and the face will no longer be the sharpest spot since it's not on the focus plane anymore. Simple as that.

The only way to do it the way you want it is with wide tracking, face/eye detection or the best is just use the focus points situated at the top of the sensor where the face will be.

Focal plane changes when you re-focus.

I know drawing is funny but that's what I wanted to show you. The red line is the focal plane.

The point you're making is spot on, however, it's worth noting again that the focal plane is actually curved, not straight as you've illustrated, and equidistant to the camera.

Question...

An online depth-of-field calculator shows that for the first image:
Focal length 39mm, f/8, that the depth of field

I think he had to be at least 6 feet from his subject when he took the photo, but the OP can confirm.

At a distance of 8 feet, the depth of field would have been:
In front of subject: 1.58 ft
Behind subject: 2.61 ft

Wouldn't that have been sufficient for Focus+Recompose to have captured sharp detail, in spite of the slight change in the focal plane?

Yes the DOF is adequate.  Second - this is not a flat field lens - there are very few flat field lenses.  Even the X80 macro has field curvature.  So the above picture showing a flat focus field is not accurate for this particular lens or most likely very few lenses.  A lens that has soft corners wide open has a curved field of focus.

Focus recompose should have worked fine in this case.

-- hide signature --

Truman
www.pbase.com/tprevatt

 Truman Prevatt's gear list:Truman Prevatt's gear list
Fujifilm X-Pro2 Fujifilm X-H1 Fujifilm XF 18mm F2 R Fujifilm XF 35mm F1.4 R Fujifilm XF 55-200mm F3.5-4.8 R LM OIS +9 more
Clive99 Contributing Member • Posts: 766
Re: Need help with focusing issue.

baobob wrote:

Not the best way for portrait, the best being a small one on the eye controlling for DOF

depending on FL and distance DOF can be very very short with eye sharp and nose or ear OOF

What you say is true in theory (as a general statement) and true for a good prime lens that can accurately focus using the smallest box. However, I've spent the last couple weeks testing the 16-80 and specifically with eye focus. I have found that the smaller (smallest and 2nd smallest single point) give unreliable results on eye focus. I actually posted a test shot of my daughter in another thread. Often the hair around her ear was in focus, not her eye - it perhaps picked a high contrast area just outside the focus box? BTW I find my 18-55 and 18-135 have similar difficulties achieving critical focus with small focus boxes. Given my findings, I'm also not sure if I would trust automatic eye focus with these lenses for critical focus.

The smallest I've used with consistent results with this lens is the 3rd smallest. I just ensure the subject is large enough in the frame that the 3rd smallest box covers the subject's eye and no other high contrast areas.

Clive

 Clive99's gear list:Clive99's gear list
Nikon 1 J5 Fujifilm X-E3 Fujifilm X-T3 Nikon 1 Nikkor VR 30-110mm f/3.8-5.6 Nikon 1 Nikkor 10mm f/2.8 +17 more
baobob
baobob Forum Pro • Posts: 13,994
Re: Need help with focusing issue.

When at close distance never focus and recompose, od course!! just use jouystick to put the AF sensor (not too big) where desired

-- hide signature --

Good judgment comes from experience
Experience comes from bad judgment

 baobob's gear list:baobob's gear list
Sony RX100 Olympus Tough TG-4 Panasonic ZS200 Fujifilm X-H1 Fujifilm X-T3 +10 more
Clive99 Contributing Member • Posts: 766
Re: Need help with focusing issue.

Nas from the Valley wrote:

baobob wrote:

Not the best way for portrait, the best being a small one on the eye controlling for DOF

depending on FL and distance DOF can be very very short with eye sharp and nose or ear OOF

The problem is not the zone focusing he used, it will be the same if he uses even the smallest focusing point. The problem is that he moves the camera down and focus plane changes.

Maybe and maybe not. I suggest he eliminate the variables. Zone being one of them. Smallest zone is too big and you have no control over what it picks. Use the smallest single point with which the lens can reliably focus. For that lens, i find that to be the 3rd smallest.

DOF can be pretty small, as Baobob said. Zone is a bad idea.

 Clive99's gear list:Clive99's gear list
Nikon 1 J5 Fujifilm X-E3 Fujifilm X-T3 Nikon 1 Nikkor VR 30-110mm f/3.8-5.6 Nikon 1 Nikkor 10mm f/2.8 +17 more
OP charley5 Senior Member • Posts: 1,771
Re: Need help with focusing issue.

You can move the focus box to where you want the face to be in the frame before you even raise the camera. If I’m shooting people I will move the the focus box to the top third of the frame for both vertical and horizontal orientations - the camera can be set to remember where you set the point for each orientation so that you don’t have keep moving it every time you flip your camera around. It can speed things up a bit - your focus box will always be in the ballpark to start.

Very true. That is what I have been doing!

 charley5's gear list:charley5's gear list
Fujifilm X-T2 Fujifilm X-H1 Fujifilm 16-55mm F2.8R LM WR
OP charley5 Senior Member • Posts: 1,771
Re: Need help with focusing issue.

Nas from the Valley wrote:

charley5 wrote:

Hi Everyone:

I am having a frustrating time with a focusing issue and I need help to solve it. I focus on the face of the person initially by pressing down the shutter half way. Then, because I want the face to be at the top third of the frame, I move the camera (with the shutter still half-way depressed) usually to chest level, and press all the way down. Their chest then is well focused in the final image but their face is not. This is really frustrating. I have lost several good shots in this way. What may be the problem? I use single point focusing and 9 squares. Thanks for any suggestions. Here are two examples:

I couldn't read all of the comments above but of course you will have a problem with focusing. It is obvious!

When you focus (in single focus mode) and hold the shutter button and move the camera you are changing focus plane! Don't expect camera to remember where exactly you focused and lock that focus spot forever!

Let me give you an example: you focus on the face which is exactly let's say 2.37 meters away from the camera and focus plane will stay on the same distance (2.37 meters) as long as you are holding the shutter button. Then you move down the camera a little bit and the face will be on let's say 2.32 meters away from the camera but the chest now will be on 2.37 meters. So the focus plane will be changed and the face will no longer be the sharpest spot since it's not on the focus plane anymore. Simple as that.

The only way to do it the way you want it is with wide tracking, face/eye detection or the best is just use the focus points situated at the top of the sensor where the face will be.

Focal plane changes when you re-focus.

I know drawing is funny but that's what I wanted to show you. The red line is the focal plane.

Very informative. Thank you.

 charley5's gear list:charley5's gear list
Fujifilm X-T2 Fujifilm X-H1 Fujifilm 16-55mm F2.8R LM WR
Truman Prevatt
Truman Prevatt Veteran Member • Posts: 9,939
Re: Adequate depth of field for Focus/Recompose

Jerry-astro wrote:

ttusteveo wrote:

Jerry-astro wrote:

Nas from the Valley wrote:

charley5 wrote:

Hi Everyone:

I am having a frustrating time with a focusing issue and I need help to solve it. I focus on the face of the person initially by pressing down the shutter half way. Then, because I want the face to be at the top third of the frame, I move the camera (with the shutter still half-way depressed) usually to chest level, and press all the way down. Their chest then is well focused in the final image but their face is not. This is really frustrating. I have lost several good shots in this way. What may be the problem? I use single point focusing and 9 squares. Thanks for any suggestions. Here are two examples:

I couldn't read all of the comments above but of course you will have a problem with focusing. It is obvious!

When you focus (in single focus mode) and hold the shutter button and move the camera you are changing focus plane! Don't expect camera to remember where exactly you focused and lock that focus spot forever!

Let me give you an example: you focus on the face which is exactly let's say 2.37 meters away from the camera and focus plane will stay on the same distance (2.37 meters) as long as you are holding the shutter button. Then you move down the camera a little bit and the face will be on let's say 2.32 meters away from the camera but the chest now will be on 2.37 meters. So the focus plane will be changed and the face will no longer be the sharpest spot since it's not on the focus plane anymore. Simple as that.

The only way to do it the way you want it is with wide tracking, face/eye detection or the best is just use the focus points situated at the top of the sensor where the face will be.

Focal plane changes when you re-focus.

I know drawing is funny but that's what I wanted to show you. The red line is the focal plane.

The point you're making is spot on, however, it's worth noting again that the focal plane is actually curved, not straight as you've illustrated, and equidistant to the camera.

Question...

An online depth-of-field calculator shows that for the first image:
Focal length 39mm, f/8, that the depth of field

I think he had to be at least 6 feet from his subject when he took the photo, but the OP can confirm.

At a distance of 8 feet, the depth of field would have been:
In front of subject: 1.58 ft
Behind subject: 2.61 ft

Wouldn't that have been sufficient for Focus+Recompose to have captured sharp detail, in spite of the slight change in the focal plane?

If you were responding to my comments, they were intended to be more generic as opposed to relating to this specific example. The larger the aperture and thinner the DOF, the more likely you are to run into issues with focus and recompose. Shooting portraits, for example, with the 56/1.2 wide open, can most definitely get one in trouble with focus and recompose as opposed to simply moving the focus point directly on your intended point of focus.

Not really. The image below was taken with a 56 f1.2 no cropping except to knock off the long dimension to make it a 4x5 aspect ratio. The distance from the camera to Dad was about 6 to 7 feet. This was taken at f1.2 at ss 1/60 using a Pro2. The center focus point was used and the focal point was right between his eyes. I then recomposed. The DOF is in fact narrow as the two design patches on the shirt front and back of the focal plane are soft, the Marine tattoo on his right bicep is soft and the "Semper" tattoo on his fore arm is quite soft.

The reason focus recompose worked on this images and really on most most lenses have a curved field of focus instead of a planer field of focus.  I know that there is all sorts of material on the internet about how one should not use focus recompose because of flat focal field.  However, it is strictly a academic exercise since very few lenses have a flat field.  If you look at the MTF of a lens - wide open you can easily tell the field of focus is not flat.  Even the  the MTF spec (on the Fuji website) is not flat.  The 56 f1.2 MTF spec is flatter than the 50 f2 although neither are bad.  Production lenses most likely will have an MTF a little worse in field curvature at a random focal distance than the spec since the spec is defined at one specific focus distance usually where the lens performs best. Any lens that is soft in the corners wide open and most lenses are - the field of focus is not flat.

If I have time I will use the joystick, but when I photography people I prefer to work fast camera to my for less than a second.  If I had held the camera up and messed with the joy stick, by the time I got ready to fire the shutter I would have been looking at Dan looking at me with his tongue sticking or or his middle finger up - or both!    What are friends for?

With people unless it is a formal portrait with the camera on a tripod, I normally focus recompose - even with the 56 wide open. If it is a formal portrait I would probably use the joy stick.

Clearly using the joy stick is one option if the shooting situations allows in some cases it should be the default.  However, one should not be afraid to use focus recompose based on the academic arguments on the Internet which are based on lenses that don't really exist in the wild.  The only true flat plane lens I nave ever owned is my Nikkor enlarging lens from my 4x5 enlarger.  But for a simple enlarging lens it was very expensive - even used.

The key is to know your camera, know your lens so you won't be surprised and you will have skill craft to make good realtime decisions.  That comes with a lot of experimentation which is difficult to come by sitting online.  

-- hide signature --

Truman
www.pbase.com/tprevatt

 Truman Prevatt's gear list:Truman Prevatt's gear list
Fujifilm X-Pro2 Fujifilm X-H1 Fujifilm XF 18mm F2 R Fujifilm XF 35mm F1.4 R Fujifilm XF 55-200mm F3.5-4.8 R LM OIS +9 more
Erik Baumgartner Senior Member • Posts: 1,428
Re: Need help with focusing issue.

charley5 wrote:

You can move the focus box to where you want the face to be in the frame before you even raise the camera. If I’m shooting people I will move the the focus box to the top third of the frame for both vertical and horizontal orientations - the camera can be set to remember where you set the point for each orientation so that you don’t have keep moving it every time you flip your camera around. It can speed things up a bit - your focus box will always be in the ballpark to start.

Very true. That is what I have been doing!

In your original post you mention focusing on the face and moving the camera to move the face to the top third of the frame, suggesting the focus box was in the center. I’m talking about moving the focus box so it’s already at the top third so you don’t have change the camera position after focusing.

 Erik Baumgartner's gear list:Erik Baumgartner's gear list
Fujifilm X-T2 Fujifilm X-T20 Fujifilm XF 35mm F1.4 R Fujifilm XF 56mm F1.2 R Fujifilm XF 50-140mm F2.8 +1 more
OP charley5 Senior Member • Posts: 1,771
Re: Need help with focusing issue.

Erik Baumgartner wrote:

charley5 wrote:

You can move the focus box to where you want the face to be in the frame before you even raise the camera. If I’m shooting people I will move the the focus box to the top third of the frame for both vertical and horizontal orientations - the camera can be set to remember where you set the point for each orientation so that you don’t have keep moving it every time you flip your camera around. It can speed things up a bit - your focus box will always be in the ballpark to start.

Very true. That is what I have been doing!

In your original post you mention focusing on the face and moving the camera to move the face to the top third of the frame, suggesting the focus box was in the center. I’m talking about moving the focus box so it’s already at the top third so you don’t have change the camera position after focusing.

Yes, I just started doing it today on my outing and it is improving the number of in-focus shots I am getting! I am about to post results from this outing and then I was going to thank everyone for the idea!

-Charles

 charley5's gear list:charley5's gear list
Fujifilm X-T2 Fujifilm X-H1 Fujifilm 16-55mm F2.8R LM WR
Jerry-astro
MOD Jerry-astro Forum Pro • Posts: 14,785
Re: Adequate depth of field for Focus/Recompose

Truman Prevatt wrote:

If you were responding to my comments, they were intended to be more generic as opposed to relating to this specific example. The larger the aperture and thinner the DOF, the more likely you are to run into issues with focus and recompose. Shooting portraits, for example, with the 56/1.2 wide open, can most definitely get one in trouble with focus and recompose as opposed to simply moving the focus point directly on your intended point of focus.

Not really. The image below was taken with a 56 f1.2 no cropping except to knock off the long dimension to make it a 4x5 aspect ratio. The distance from the camera to Dad was about 6 to 7 feet. This was taken at f1.2 at ss 1/60 using a Pro2. The center focus point was used and the focal point was right between his eyes. I then recomposed. The DOF is in fact narrow as the two design patches on the shirt front and back of the focal plane are soft, the Marine tattoo on his right bicep is soft and the "Semper" tattoo on his fore arm is quite soft.

The reason focus recompose worked on this images and really on most most lenses have a curved field of focus instead of a planer field of focus. I know that there is all sorts of material on the internet about how one should not use focus recompose because of flat focal field. However, it is strictly a academic exercise since very few lenses have a flat field. If you look at the MTF of a lens - wide open you can easily tell the field of focus is not flat. Even the the MTF spec (on the Fuji website) is not flat. The 56 f1.2 MTF spec is flatter than the 50 f2 although neither are bad. Production lenses most likely will have an MTF a little worse in field curvature at a random focal distance than the spec since the spec is defined at one specific focus distance usually where the lens performs best. Any lens that is soft in the corners wide open and most lenses are - the field of focus is not flat.

If I have time I will use the joystick, but when I photography people I prefer to work fast camera to my for less than a second. If I had held the camera up and messed with the joy stick, by the time I got ready to fire the shutter I would have been looking at Dan looking at me with his tongue sticking or or his middle finger up - or both! What are friends for?

With people unless it is a formal portrait with the camera on a tripod, I normally focus recompose - even with the 56 wide open. If it is a formal portrait I would probably use the joy stick.

Clearly using the joy stick is one option if the shooting situations allows in some cases it should be the default. However, one should not be afraid to use focus recompose based on the academic arguments on the Internet which are based on lenses that don't really exist in the wild. The only true flat plane lens I nave ever owned is my Nikkor enlarging lens from my 4x5 enlarger. But for a simple enlarging lens it was very expensive - even used.

The key is to know your camera, know your lens so you won't be surprised and you will have skill craft to make good realtime decisions. That comes with a lot of experimentation which is difficult to come by sitting online.

Umm, whatever, Truman.  Plain and simple if you shoot two people standing side by side at a large aperture, and focus on one person, you can run the risk of having the other one slightly OOF.  And the reason for that is exactly the point you made, which is that you have a curved rather than linear field.  Thus, the distance from the camera of one person may be a touch further than the other in spite of the fact that they're standing side by side.  So, please explain how your explanation differs from that (other than being a lot longer) or what I said earlier?

-- hide signature --

Jerry-Astro
Fujifilm X Forum Co-Mod

 Jerry-astro's gear list:Jerry-astro's gear list
Fujifilm X-H1 Carl Zeiss Touit 2.8/12 Fujifilm XF 10-24mm F4 R OIS Fujifilm 16-55mm F2.8R LM WR Fujifilm XF 100-400mm F4.5-5.6 OIS WR +1 more
Keyboard shortcuts:
FForum MMy threads