Question about focus peaking: Nikon Z Mirrorless Talk Forum: Digital Photography Review (2024)

I am a fairly recent Z50 owner and have always enjoyed using various manual focus lenses on previous ILC cameras. Having never owned a camera with focus peaking I was excited to try it out. My experiences have been a bit disappointing. I have a smattering of manual lenses, old F lenses, Russian, East German, Chinese, etc. what I am finding is that no matter how sensitive I have peaking set, there is poor accuracy. For instance I had a Meike 25 f/1.8 lens on my Z50, set to f/4 and peaking showed that there were things in focus in roughly a 2ft depth which seems like a lot of DOF for the settings and the subsequent shot showed the thing I thought I was focusing on was definitely not in focus. I should mention that this was a low light situation which is exactly the situation I would want to use this. Am I totally missing the point of focus peaking? I should add that I have no problem focusing manually without peaking on and I have taken many, many successful manual focus shots over the years with my V1s which of course do not have focus peaking.

3

Paul Pasco wrote:

I am a fairly recent Z50 owner and have always enjoyed using various manual focus lenses on previous ILC cameras. Having never owned a camera with focus peaking I was excited to try it out. My experiences have been a bit disappointing. I have a smattering of manual lenses, old F lenses, Russian, East German, Chinese, etc. what I am finding is that no matter how sensitive I have peaking set, there is poor accuracy. For instance I had a Meike 25 f/1.8 lens on my Z50, set to f/4 and peaking showed that there were things in focus in roughly a 2ft depth which seems like a lot of DOF for the settings and the subsequent shot showed the thing I thought I was focusing on was definitely not in focus. I should mention that this was a low light situation which is exactly the situation I would want to use this. Am I totally missing the point of focus peaking? I should add that I have no problem focusing manually without peaking on and I have taken many, many successful manual focus shots over the years with my V1s which of course do not have focus peaking.

Can you post a sample please? Even better, if you could take one of your lcd while peaking is on, and then the actual picture it took so that we can compare the peaking to the results.

When you say that the DoF was roughly 2 ft in depth, my question is: were you looking at the thing you wanted in focus or were you looking at the surroundings?

Sometimes, surroundings can be misleading. For example, if you are taking a full body portrait of someone standing while you are also standing, the floor at that distance (or feet of the person) should not be in focus, since this would be further than the person's head. This can be particularly true for older / simpler lenses that might exhibit more field curvature.

I find focus peaking on my Z6 to be generally adequate--I usually quickly move the focus ring back and forth a few times to get a better feel for where the focus is. If I want critical focus, I use magnification as well--I have this set to one of my function buttons, so it's literally 1 easy button to magnify.

2

Maybe a short lecture on how focus peaking works is on order:

No AF system or other auxiliary subsystem or unit is involved in it whatsoever. The camera's CPU analyzes the image read straight from the sensor and looks for the sharpest area(s), which it then highlights.

There are different ways of determining the sharpest area. The simplest one is essentially to compare the difference between the R/G/B values of adjacent pixels: the bigger the cumulative difference, the sharper a transition exists between these pixels. Manual focus, AF, old lens, new lens, American, Japanese, Russian - all of these aspects are totally irrelevant here. The sharpest areas in the resulting image are the only thing that matters.

Regardless of which exact algorithm the camera's focus peaking uses, it CANNOT be inaccurate in a sense of showing the wrong focus/sharpness point. The algorithm CAN show a smaller area in focus than there actually is, or it CAN show a larger area in focus than there actually is, but it can never show the WRONG area. If nothing is really sharp in your image, it might show an area that surprises you, but, trust me, that is still the sharpest area in the image.

As I hope we all understand, 'smaller area' respectively 'larger area' are misleading terms here anyway, as technically, only a single point in your image can be in full focus, meaning at maximum sharpness. By choosing the algorithm, the camera maker sort of decides how little or how much deviation from that sharpest point the camera still considers 'in focus' and thus marks by flashing via the focus peaking function.

lokatz wrote:

Maybe a short lecture on how focus peaking works is on order:

No AF system or other auxiliary subsystem or unit is involved in whatsoever. Instead, the camera's CPU analyzes the image read straight from the sensor and looks for the sharpest area(s), which it then flashes.

There are different ways of determining the sharpest area. The simplest one is essentially to compare the difference between the R/G/B values of adjacent pixels: the bigger the cumulative difference, the sharper a transition exists between these pixels. Manual focus, AF, old lens, new lens, American, Japanese, Russian - all of these aspects are totally irrelevant here.

Regardless of which exact algorithm the camera's focus peaking uses, it CANNOT be inaccurate in a sense of showing the wrong focus/sharpness point. The algorithm CAN show a smaller area in focus than there actually is, or it CAN show a larger area in focus than there actually is, but it can never show the WRONG area.

As I hope we all understand, 'smaller area' respectively 'larger area' are misleading terms here anyway, as technically, only a single point in your image can be in full focus, meaning at maximum sharpness. By choosing the algorithm, the camera maker sort of decides how little or how much deviation from that sharpest point the camera still considers 'in focus' and thus marks by flashing via the focus peaking function.

Just wanted to point out: This isn't always how focus peaking works. Sometimes peaking can be based on the result of analysis of the PDAF (half masked) pixels, with the only difference between this and PDAF being it doesn't actually drive an AF motor. Some mirrorless cameras even allow you to see this in the EVF, such as Fuji's "digital split image," which shows the composite image from half of the the half-masked pixels next to the image from the other set of half-masked pixels). In practice, not too dissimilar from how split prism focus screens worked on SLRs. In other cases, the contrast detection you mentioned will be used; and sometimes, some combination of both will be leveraged.

I've never studied which specific system or methods the Z's use for peaking. I wouldn't be surprised if they primarily or exclusively use contrast detection (analyzing blur / transitions), but I wanted to point out that it's not necessarily all contrast detection.

Another thing I wanted to point out is that it is certainly possible for peaking to show the wrong areas in focus. One example: a lens with harsh bokeh. If you have harsh bokeh, this can appear as being a rapid transition of edges in focus. So it does depend quite a bit on the methods & algorithms the cameras use.

Finally, there isn't technically only a single point in "full focus." "Full focus" is typically on an entire plane, or a spherical sector ("curved plane"?), or other volumes. The typical commonality is in taking a 2D plane that's parallel to the surface of the lens and then "curving" it in the 3rd dimension in various ways, depending on the optics. Anything that intersects this should be considered to be in "full focus," with various distances from this (along the Z axis) being within the DoF but not in "full focus."

So all in all, it's pretty complicated--and that's just the technicalities, before accounting for potential human error and various circ*mstances (such as subjects, lens properties, etc.). So I'd be interested in seeing examples, especially if they compare the focus peaked areas to the actual results.

1

Paul Pasco wrote:

I am a fairly recent Z50 owner and have always enjoyed using various manual focus lenses on previous ILC cameras. Having never owned a camera with focus peaking I was excited to try it out. My experiences have been a bit disappointing. I have a smattering of manual lenses, old F lenses, Russian, East German, Chinese, etc. what I am finding is that no matter how sensitive I have peaking set, there is poor accuracy. For instance I had a Meike 25 f/1.8 lens on my Z50, set to f/4 and peaking showed that there were things in focus in roughly a 2ft depth which seems like a lot of DOF for the settings and the subsequent shot showed the thing I thought I was focusing on was definitely not in focus.

I've seen it recommended to use the lowest sensitivity setting (1 I think) for more precise focus.

I should mention that this was a low light situation which is exactly the situation I would want to use this. Am I totally missing the point of focus peaking? I should add that I have no problem focusing manually without peaking on and I have taken many, many successful manual focus shots over the years with my V1s which of course do not have focus peaking.

Although I'm not (yet) a Z user, I trust Paul's expertise from his involvement in the Nikon 1 forum, and I know already that we we both don't like zooming in to focus because of inaccuracies while re-composing. (Zooming in to focus is possible only with the FT1 and Nikon F mount lenses, not dumb adapters.)

I therefore think the problem is the registration with other mounts so that they allow for infinity and beyond (let's call them "Buzz Lightyear" adapters! 🤣) mainly to account for heat differences with telephoto lenses. In other words, the camera might think the lens is in focus, when it was actually matched to another distance.

I realize that this sounds a bit illogical since the correction happens in-camera and the lens shouldn't be the culprit, but why do lenses need to be tuned for focus on a DSLR? It's all calibration accuracy, and Nikon does have a reputation of supporting only its own products. So to me, this argument is logical.

2

Parry Johnson wrote:

Although I'm not (yet) a Z user, I trust Paul's expertise from his involvement in the Nikon 1 forum, and I know already that we we both don't like zooming in to focus because of inaccuracies while re-composing. (Zooming in to focus is possible only with the FT1 and Nikon F mount lenses, not dumb adapters.)

I therefore think the problem is the registration with other mounts so that they allow for infinity and beyond (let's call them "Buzz Lightyear" adapters! 🤣) mainly to account for heat differences with telephoto lenses. In other words, the camera might think the lens is in focus, when it was actually matched to another distance.

I realize that this sounds a bit illogical since the correction happens in-camera and the lens shouldn't be the culprit, but why do lenses need to be tuned for focus on a DSLR? It's all calibration accuracy, and Nikon does have a reputation of supporting only its own products. So to me, this argument is logical.

Lenses need to be tuned for focus on a DSLR primarily because:

• They don't focus at the same aperture as is used for the image (focus shift)
• They don't focus at the same location (they focus at the AF module, which is at the bottom of the camera. They don't use the imaging sensor, which is at the back of the camera).

These are probably the most dominant factors and reasons for calibration & inaccuracy on DSLRs.

But critically: the mirrorless cameras including the Z don't operate like this. So these major factors would be irrelevant to the Z's. Meaning the registration distances also shouldn't make any difference whatsoever with "dumb" adapters.

BTW, I have both Nikon 1 & Z cameras. The manual focus experience is very different, including in magnification. On the Z cameras, you can magnify at any point, using any lens, with no need to focus & recompose. And this works with peaking as well (peaking + magnification).

4

CAUTION - focus peaking highlighting compared to what is within the zone of good sharpness are not exactly the same.

Peaking gives an indication (as already explained by others) of what has good contrast.

Zooming in to 50%, 100% or 200% (menu choices) which I do using the Fn2 button helps much better determine where focus should be placed.

There is always more depth of field behind the point of critical focus than in front.

With wide angles there is usually a lot more dof behind the point of best sharpness than in front

Not mentioned so far is focus peaking only highlights detail parallel to the short dimension of the frame. It does not highlight detail parallel to the long dimension of the frame.

1

Thanks to all of the responders, I didn’t mean to “hit and run”, just been busy.

beatboxa wrote:

Paul Pasco wrote:

I am a fairly recent Z50 owner and have always enjoyed using various manual focus lenses on previous ILC cameras. Having never owned a camera with focus peaking I was excited to try it out. My experiences have been a bit disappointing. I have a smattering of manual lenses, old F lenses, Russian, East German, Chinese, etc. what I am finding is that no matter how sensitive I have peaking set, there is poor accuracy. For instance I had a Meike 25 f/1.8 lens on my Z50, set to f/4 and peaking showed that there were things in focus in roughly a 2ft depth which seems like a lot of DOF for the settings and the subsequent shot showed the thing I thought I was focusing on was definitely not in focus. I should mention that this was a low light situation which is exactly the situation I would want to use this. Am I totally missing the point of focus peaking? I should add that I have no problem focusing manually without peaking on and I have taken many, many successful manual focus shots over the years with my V1s which of course do not have focus peaking.

Can you post a sample please? Even better, if you could take one of your lcd while peaking is on, and then the actual picture it took so that we can compare the peaking to the results.

I will try to do that time permitting.

When you say that the DoF was roughly 2 ft in depth, my question is: were you looking at the thing you wanted in focus or were you looking at the surroundings?

What I meant was peaking was showing me that my subject (dog) was highlighted and objects up to 3 feet behind him were also highlighted and when I took the shot the dog was totally OOF and the objects behind were well focused. I should mention that the dog was sitting still. My mistake apparently was assuming anything highlighted would be in focus.

Sometimes, surroundings can be misleading. For example, if you are taking a full body portrait of someone standing while you are also standing, the floor at that distance (or feet of the person) should not be in focus, since this would be further than the person's head. This can be particularly true for older / simpler lenses that might exhibit more field curvature.

I find focus peaking on my Z6 to be generally adequate--I usually quickly move the focus ring back and forth a few times to get a better feel for where the focus is. If I want critical focus, I use magnification as well--I have this set to one of my function buttons, so it's literally 1 easy button to magnify.

beatboxa wrote:

Parry Johnson wrote:

Although I'm not (yet) a Z user, I trust Paul's expertise from his involvement in the Nikon 1 forum, and I know already that we we both don't like zooming in to focus because of inaccuracies while re-composing. (Zooming in to focus is possible only with the FT1 and Nikon F mount lenses, not dumb adapters.)

I therefore think the problem is the registration with other mounts so that they allow for infinity and beyond (let's call them "Buzz Lightyear" adapters! 🤣) mainly to account for heat differences with telephoto lenses. In other words, the camera might think the lens is in focus, when it was actually matched to another distance.

I realize that this sounds a bit illogical since the correction happens in-camera and the lens shouldn't be the culprit, but why do lenses need to be tuned for focus on a DSLR? It's all calibration accuracy, and Nikon does have a reputation of supporting only its own products. So to me, this argument is logical.

Lenses need to be tuned for focus on a DSLR primarily because:

• They don't focus at the same aperture as is used for the image (focus shift)
• They don't focus at the same location (they focus at the AF module, which is at the bottom of the camera. They don't use the imaging sensor, which is at the back of the camera).

These are probably the most dominant factors and reasons for calibration & inaccuracy on DSLRs.

But critically: the mirrorless cameras including the Z don't operate like this. So these major factors would be irrelevant to the Z's. Meaning the registration distances also shouldn't make any difference whatsoever with "dumb" adapters.

BTW, I have both Nikon 1 & Z cameras. The manual focus experience is very different, including in magnification. On the Z cameras, you can magnify at any point, using any lens, with no need to focus & recompose. And this works with peaking as well (peaking + magnification).

I think what Parry means by “focus and recompose” is that in the act of magnifying and switching back to full screen there is the possibility of camera movement. I myself find it disconcerting and maybe I just need to practice more to be comfortable. My experience goes back to film days with TLRs and SLRs and the only magnifier I had then was the flip up one on my Yashica Mat TLR!

3

Paul Pasco wrote:

What I meant was peaking was showing me that my subject (dog) was highlighted and objects up to 3 feet behind him were also highlighted and when I took the shot the dog was totally OOF and the objects behind were well focused. I should mention that the dog was sitting still. My mistake apparently was assuming anything highlighted would be in focus.

A couple of suggestions: Make sure you understand the focus peaking sensitivity setting. I find it confusing. High sensitivity means a great depth with appear in focus, in other words less precision. Low sensitivity means a more precise, limited depth with be highlighted. As for technique, I find it necessary to make small focus adjustments and watch the highlights sweep back and forth in the image. This helps me better judge where the plane of maximum focus is located. If the subject is stationary, zooming in (as others have already mentioned) can help a lot too.

Paul Pasco wrote:

beatboxa wrote:

Paul Pasco wrote:

I am a fairly recent Z50 owner and have always enjoyed using various manual focus lenses on previous ILC cameras. Having never owned a camera with focus peaking I was excited to try it out. My experiences have been a bit disappointing. I have a smattering of manual lenses, old F lenses, Russian, East German, Chinese, etc. what I am finding is that no matter how sensitive I have peaking set, there is poor accuracy. For instance I had a Meike 25 f/1.8 lens on my Z50, set to f/4 and peaking showed that there were things in focus in roughly a 2ft depth which seems like a lot of DOF for the settings and the subsequent shot showed the thing I thought I was focusing on was definitely not in focus. I should mention that this was a low light situation which is exactly the situation I would want to use this. Am I totally missing the point of focus peaking? I should add that I have no problem focusing manually without peaking on and I have taken many, many successful manual focus shots over the years with my V1s which of course do not have focus peaking.

Can you post a sample please? Even better, if you could take one of your lcd while peaking is on, and then the actual picture it took so that we can compare the peaking to the results.

I will try to do that time permitting.

When you say that the DoF was roughly 2 ft in depth, my question is: were you looking at the thing you wanted in focus or were you looking at the surroundings?

What I meant was peaking was showing me that my subject (dog) was highlighted and objects up to 3 feet behind him were also highlighted and when I took the shot the dog was totally OOF and the objects behind were well focused. I should mention that the dog was sitting still. My mistake apparently was assuming anything highlighted would be in focus.

This is precisely why I'd love to see the example/comparison. I suspect that the dog may not have been in the "center" of the focus peaking range for whichever setting you used--and that this is exacerbated by some less-than-intuitive factors such as distance vs. angles.

This reminds me of a thread from several years ago:

• https://www.dpreview.com/forums/post/58928707

I went back and manually shaded in red what I thought appeared to be in focus. And the reason I bring it up is to reiterate that "dome" concept I alluded to earlier.

Remember that distances will be the same in a dome--so it is possible totake a picture of your dog from a position where both your dog's head and the floor behind your dog would be in focus at the same time.

Also remember that focus peaking is based on a spectrum. You can narrow this spectrum through the sensitivity setting; but you also should want to ensure that you are as close to the center of this spectrum as possible for accurate focusing.

Ernie Misner wrote:

Paul Pasco wrote:

I am a fairly recent Z50 owner and have always enjoyed using various manual focus lenses on previous ILC cameras. Having never owned a camera with focus peaking I was excited to try it out. My experiences have been a bit disappointing. I have a smattering of manual lenses, old F lenses, Russian, East German, Chinese, etc. what I am finding is that no matter how sensitive I have peaking set, there is poor accuracy. For instance I had a Meike 25 f/1.8 lens on my Z50, set to f/4 and peaking showed that there were things in focus in roughly a 2ft depth which seems like a lot of DOF for the settings and the subsequent shot showed the thing I thought I was focusing on was definitely not in focus.

I've seen it recommended to use the lowest sensitivity setting (1 I think) for more precise focus.

I don’t know about precision but anything more than the lowest is very distracting.

I should mention that this was a low light situation which is exactly the situation I would want to use this. Am I totally missing the point of focus peaking? I should add that I have no problem focusing manually without peaking on and I have taken many, many successful manual focus shots over the years with my V1s which of course do not have focus peaking.

1

Paul Pasco wrote:

I am a fairly recent Z50 owner and have always enjoyed using various manual focus lenses on previous ILC cameras. Having never owned a camera with focus peaking I was excited to try it out. My experiences have been a bit disappointing. I have a smattering of manual lenses, old F lenses, Russian, East German, Chinese, etc. what I am finding is that no matter how sensitive I have peaking set, there is poor accuracy. For instance I had a Meike 25 f/1.8 lens on my Z50, set to f/4 and peaking showed that there were things in focus in roughly a 2ft depth which seems like a lot of DOF for the settings and the subsequent shot showed the thing I thought I was focusing on was definitely not in focus. I should mention that this was a low light situation which is exactly the situation I would want to use this. Am I totally missing the point of focus peaking? I should add that I have no problem focusing manually without peaking on and I have taken many, many successful manual focus shots over the years with my V1s which of course do not have focus peaking.

Focus peaking does nothing to show you the DOF. It does not know your aperture for starters, or the focal length, or focus distance, or what COC is acceptable for you.. For DOF you need to consult a calculator (on your smartphone for example), or the DOF markers on an old handfocused lens.

It just highlights all the places that the peaking thinks are exactly in focus, based on contrast.

In principle it works, just for focus. I would say, try it out. Take along a measuring tape, put an object at 1 m, focus with peaking, and see what the distance ring of your lens is telling you. And that 1 m is not from the front of the lens, but from a stripe with a circle through it on the camerabody, that tells you where the sensor is located.

And I use blue for peaking, most contrast with reddish faces and greenish foliage. Focusing on the blue sky is not going to work anyway

Succes!

Boudewijn van der Drift wrote:

Paul Pasco wrote:

I am a fairly recent Z50 owner and have always enjoyed using various manual focus lenses on previous ILC cameras. Having never owned a camera with focus peaking I was excited to try it out. My experiences have been a bit disappointing. I have a smattering of manual lenses, old F lenses, Russian, East German, Chinese, etc. what I am finding is that no matter how sensitive I have peaking set, there is poor accuracy. For instance I had a Meike 25 f/1.8 lens on my Z50, set to f/4 and peaking showed that there were things in focus in roughly a 2ft depth which seems like a lot of DOF for the settings and the subsequent shot showed the thing I thought I was focusing on was definitely not in focus. I should mention that this was a low light situation which is exactly the situation I would want to use this. Am I totally missing the point of focus peaking? I should add that I have no problem focusing manually without peaking on and I have taken many, many successful manual focus shots over the years with my V1s which of course do not have focus peaking.

Focus peaking does nothing to show you the DOF. It does not know your aperture for starters, or the focal length, or focus distance, or what COC is acceptable for you.. For DOF you need to consult a calculator (on your smartphone for example), or the DOF markers on an old handfocused lens.

I didn’t say it shows DOF, on the contrary I said it seems to show things in focus that can’t or shouldn’t be! That’s why I asked the question.

It just highlights all the places that the peaking thinks are exactly in focus, based on contrast.

In principle it works, just for focus. I would say, try it out. Take along a measuring tape, put an object at 1 m, focus with peaking, and see what the distance ring of your lens is telling you. And that 1 m is not from the front of the lens, but from a stripe with a circle through it on the camerabody, that tells you where the sensor is located.

And I use blue for peaking, most contrast with reddish faces and greenish foliage. Focusing on the blue sky is not going to work anyway

Succes!

Boudewijn van der Drift wrote:

Focus peaking does nothing to show you the DOF. It does not know your aperture for starters, or the focal length, or focus distance, or what COC is acceptable for you.. For DOF you need to consult a calculator (on your smartphone for example), or the DOF markers on an old handfocused lens.

It just highlights all the places that the peaking thinks are exactly in focus, based on contrast.

These comments may not be 100% accurate.

As you point out what is highlighted in peaking and actual depth of field are not the same thing.

With ML, using focus peaking the camera does "knows" the aperture - between wide open and f5.6 - because ML uses the shooting aperture (to a minimum of f5.6) for viewing either through the viewfinder or on the rear monitor - and for focus peaking.

With the camera on a tripod and a subject containinga tiled roof at 45° altering the aperture widens or narrows the zone highlighted by focus peaking.

I consider focus peaking at the lowest setting as a useful aid in deciding what might be within the depth of field at some unspecified aperture, but prefer to zoom in on the image for clarification as to what might be sharp in a final image.

Taking an actual image and zooming in on the rear monitor can be a useful aid in learning what is likely to be within a zone of depth of field