Looking for true focus bracketing - not focus stacking

jonby

Leading Member
Messages
665
Solutions
2
Reaction score
522
Location
UK
I’m looking for a camera system with a true focus bracketing rather than focus stacking feature. I currently mainly use Canon cameras and their ‘focus bracketing’ is definitely aimed at focus stacking rather than bracketing, and does not meet my needs.

In a nutshell, what I’m looking for is something which offers a sequence of three images - one at the current focus distance, one focused very slightly further away and one very slightly nearer, with the focus distance shifted back to its original position following the sequence. I ideally don’t want to stack images - I want to increase my chances of getting optimal focus throughout the scene in one of the shots.

For a bit of context, I’m a static scene photographer with many years’ experience. Most of the time I want the entire scene or most of the scene sharp as possible, though not always. I mostly work with smallish apertures. Most of the time I can get focus close to optimal, but this is not trivial when you’re pushing the boundaries of the depth of field, or don’t have anything at the optimal distance to focus on, and especially with the field curvature of most lenses being less than flat, and often unpredictable. I want insurance that I will have a shot where focus is optimal, so I can concentrate on getting everything else right. The depth of field in each shot needs to mostly overlap, but with a slight bias forwards or backwards. In previous decades, when lenses had well-marked focusing scales and tight focusing actions, I would commonly bracket focus manually using the distance scale on the lens, but that’s not practical now with focus-by-wire lenses.

Canon’s ‘focus bracketing’ feature fails to meet my needs in a number of ways. The biggest problem is that even with the focus increments set to the minimum, the distance steps are far too large. The depth of field in each shot barely overlaps, and the distance steps only get wider as you stop down. The second problem is that the focus distance is not reset to its original position after the sequence - it is just left wherever it ends up - meaning if I want to take a second sequence (eg. because exposure or composition wasn’t perfect) I have to re-focus. The third problem is that typically, none of the shots have the focus set exactly where I set it - the closest one usually being visibly front- or back- focused relative to what I chose, because the focus shifts are so big, meaning that I’m actually less likely to get a shot with optimal focus. Ideally, I would prefer the first shot to be the current focus distance, then one further and one closer, similar to how exposure bracketing works.

Ironically, before the whole focus stacking ‘craze’ took hold, many of Canon’s Powershot cameras worked in exactly the way that I want. I have an original G1X and the focus bracketing system works well for me. Probably 85% of the time the first shot of the sequence will be best; in the other 15% of cases, one of the others will be better and I’ll use that. Occasionally I might combine two of the images, but this is a last resort. It has been one of the best cameras I’ve ever used in terms of getting things optimally sharp, but it has its limitations, including a slow continuous shooting rate. I’d like to get this same consistency in an interchangeable lens camera which has improved IQ and flexibility, for occasions when the G1X doesn’t work.

I’ve tried this with the R5, R6 and R10, with a number of RF lenses. The problems I describe above have been consistent in all cases.

So does anyone know of a camera which works in the way that I describe? Or closer than the Canon system? Or do all systems work the same now? Would I find the same issues with Sony and Nikon, what about Olympus/OM? The most important thing is to have much finer increments, such that the depth of field in each shot substantially overlaps, and is only a very small shift in focus.Thanks for in advance for your consideration.
 
Last edited:
Can you give us an idea of the magnitude of the depth of field in the static scenes (a few millimeters, a meter, 1 kilometer)? Or a sample image?

My very first thought was a focusing rail, but that won't be practical if you're shooting landscapes.
 
Olympus (at least OM1m2) will focus bracket and create a composite, but will also save the individual source images

Check out page 250 here https://learnandsupport.getolympus.com/sites/default/files/media/files/2024/01/OM-1Mk2_ENU_00.pdf

You can adjust the number of shots and the focus differential (1 thru 10), but I can't find any information about what the focus differential means.

It works with specific lenses https://learnandsupport.getolympus....s-are-compatible-with-the-focus-stacking-mode

A bit more info here https://learnandsupport.getolympus....ips/macro/focus-stacking-bracketing-with-om-d

--

Sherm

Sherms flickr page

P950 album

P900 album RX10iv album
OM1.2 150-600 album
 
Google: Jim Kasson focus bracketing. Back around 2020, on his blog, he looked at several focus bracketing systems, including Nikon and Fuji GFX, don't remember which, if any, others. IIRC, the GFX was the only one for which he thought the step sizes to be close-enough together. IIRC, his interest in bracketing was for ultimately stacking, so his investigations probably didn't address operating the camera as you desire, but anything he discovered about step sizes should still be relevant to your intended use.

Jim is also a moderator at the Medium Format forum here at DPR. Perhaps post or PM him there, if you don't get answers here or via Googled results.
 
Can you give us an idea of the magnitude of the depth of field in the static scenes (a few millimeters, a meter, 1 kilometer)? Or a sample image?

My very first thought was a focusing rail, but that won't be practical if you're shooting landscapes.
It varies, but objects in the scene may range from around 4m to infinity. No - a focusing rail will not be practical. I'm generally shooting handheld also.
 
Google: Jim Kasson focus bracketing. Back around 2020, on his blog, he looked at several focus bracketing systems, including Nikon and Fuji GFX, don't remember which, if any, others. IIRC, the GFX was the only one for which he thought the step sizes to be close-enough together. IIRC, his interest in bracketing was for ultimately stacking, so his investigations probably didn't address operating the camera as you desire, but anything he discovered about step sizes should still be relevant to your intended use.

Jim is also a moderator at the Medium Format forum here at DPR. Perhaps post or PM him there, if you don't get answers here or via Googled results.
Thanks for this tip. This looks interesting and something I will read when I get more time (will take some brain power to digest), but a quick scan suggests that I don't think it will resolve many aspects of my question. As you say, it seems to target stacking-type usage rather than the purpose I have in mind.
 
Olympus (at least OM1m2) will focus bracket and create a composite, but will also save the individual source images

Check out page 250 here https://learnandsupport.getolympus.com/sites/default/files/media/files/2024/01/OM-1Mk2_ENU_00.pdf

You can adjust the number of shots and the focus differential (1 thru 10), but I can't find any information about what the focus differential means.

It works with specific lenses https://learnandsupport.getolympus....s-are-compatible-with-the-focus-stacking-mode

A bit more info here https://learnandsupport.getolympus....ips/macro/focus-stacking-bracketing-with-om-d
Thanks for that. The page you mention covers the focus stacking feature. However, a search of the manual revealed that the camera has a separate focus bracketing feature, which is interesting. However, both features appear to have the same options and very similar description, suggesting that the only difference may be that the stacking option will output a stacked JPEG, while the bracketing feature will not. However, I would need someone with experience of using them to confirm. It's quite possible that the distance steps may be small enough to be of use for my purposes, but this can't be determined from this manual alone. Maybe I should post something on the M43 forum.
 
Olympus (at least OM1m2) will focus bracket and create a composite, but will also save the individual source images

Check out page 250 here https://learnandsupport.getolympus.com/sites/default/files/media/files/2024/01/OM-1Mk2_ENU_00.pdf

You can adjust the number of shots and the focus differential (1 thru 10), but I can't find any information about what the focus differential means.

It works with specific lenses https://learnandsupport.getolympus....s-are-compatible-with-the-focus-stacking-mode

A bit more info here https://learnandsupport.getolympus....ips/macro/focus-stacking-bracketing-with-om-d
Thanks for that. The page you mention covers the focus stacking feature. However, a search of the manual revealed that the camera has a separate focus bracketing feature, which is interesting. However, both features appear to have the same options and very similar description, suggesting that the only difference may be that the stacking option will output a stacked JPEG, while the bracketing feature will not. However, I would need someone with experience of using them to confirm. It's quite possible that the distance steps may be small enough to be of use for my purposes, but this can't be determined from this manual alone. Maybe I should post something on the M43 forum.
The menus have changed over time. At least on the OM1m2, I don't see separate options for bracketing and stacking. You always get the source files (which can be raw or jpg) and you always get the stacked composite.

I've not seen anything there relevant to your question in the past 6 months or so, but M43 forum might be a good place for a post. The macro forum might also be worth considering.

Jim Kasson's article suggested that at least for other cameras, the step distance can be very small.

--

Sherm

Sherms flickr page

P950 album

P900 album RX10iv album
OM1.2 150-600 album
 
Take this as grain of salt. IIRC, isn't there something like a pre-MF and you can recall it by pressing an assigned fn button. I think I saw it on one of the YT videos. Can't remember the camera though.

I might also be full of it, and it might not be what you're looking for.
 
Last edited:
Have a look at Panasonic. On my ( now getting a bit old ) G9 focus bracketing can be set up as an independent function. It can certainly be configured to do 3 steps 0/-/+. The potential snag is step size, which is an arbtrary parameter with no indication of how far focus is moved. The only clue from the manual is that the degree of movement is somehow scaled with perceived subject distance.
 
Google: Jim Kasson focus bracketing. Back around 2020, on his blog, he looked at several focus bracketing systems, including Nikon and Fuji GFX, don't remember which, if any, others. IIRC, the GFX was the only one for which he thought the step sizes to be close-enough together. IIRC, his interest in bracketing was for ultimately stacking, so his investigations probably didn't address operating the camera as you desire, but anything he discovered about step sizes should still be relevant to your intended use.

Jim is also a moderator at the Medium Format forum here at DPR. Perhaps post or PM him there, if you don't get answers here or via Googled results.
I have now read the Jim Kasson article 'How focus bracketing systems work' in more detail, if that's the one you meant.

My key takeaway relating to my original question, if I have understood it correctly, is that there's a big difference in the smallest focus shift step available on the Fuji and Nikon cameras tested, with the fuji (GFX 50 and 100) being much smaller than the Nikon (Z7), by a factor of at least 6.

It sounds like the smallest step on the Nikon roughly equates to what I have experienced with Canon R cameras - ie. not small enough, whereas the Fuji probably is small enough. Unfortunately, the Fuji GFX line are not in the running for my purposes. If Fuji use a similarly small step on their X series then I could look at those, but of course there's no guarantee that that will be the case without further info.
 
Olympus (at least OM1m2) will focus bracket and create a composite, but will also save the individual source images

Check out page 250 here https://learnandsupport.getolympus.com/sites/default/files/media/files/2024/01/OM-1Mk2_ENU_00.pdf

You can adjust the number of shots and the focus differential (1 thru 10), but I can't find any information about what the focus differential means.

It works with specific lenses https://learnandsupport.getolympus....s-are-compatible-with-the-focus-stacking-mode

A bit more info here https://learnandsupport.getolympus....ips/macro/focus-stacking-bracketing-with-om-d
Thanks for that. The page you mention covers the focus stacking feature. However, a search of the manual revealed that the camera has a separate focus bracketing feature, which is interesting. However, both features appear to have the same options and very similar description, suggesting that the only difference may be that the stacking option will output a stacked JPEG, while the bracketing feature will not. However, I would need someone with experience of using them to confirm. It's quite possible that the distance steps may be small enough to be of use for my purposes, but this can't be determined from this manual alone. Maybe I should post something on the M43 forum.
The menus have changed over time. At least on the OM1m2, I don't see separate options for bracketing and stacking. You always get the source files (which can be raw or jpg) and you always get the stacked composite.
According to the manual p278, focus bracketing comes under Camera2 > Bracketing, whereas the stacking comes under Camera2 > Computational modes.
I've not seen anything there relevant to your question in the past 6 months or so, but M43 forum might be a good place for a post. The macro forum might also be worth considering.
I did find this thread from 2016 by Richard Turton. Not sure I've got my head around it fully, but it seems to suggest that the smallest step sizes on Olympus (EM-1) are smaller than what I am getting on Canon R bodies, so worth further consideration.
Jim Kasson's article suggested that at least for other cameras, the step distance can be very small.
Yes I've read it fully now, and the GFX cameras tested would appear to have very fine steps, which would probably meet my needs. Unfortunately these cameras are not in the running for my main purposes (need something smaller, lighter and cheaper), but the X series might be worth looking into.
 
Last edited:
Have a look at Panasonic. On my ( now getting a bit old ) G9 focus bracketing can be set up as an independent function. It can certainly be configured to do 3 steps 0/-/+. The potential snag is step size, which is an arbtrary parameter with no indication of how far focus is moved. The only clue from the manual is that the degree of movement is somehow scaled with perceived subject distance.
Thanks - yes I had a look and you're right - you can configure it to start with current focus position and then shoot one closer and further. This is a good start. There's no indication in the manual as to whether focus is re-set to the original position following the sequence though. The size of the steps is also unknown unless I can find out more.
 
Take this as grain of salt. IIRC, isn't there something like a pre-MF and you can recall it by pressing an assigned fn button. I think I saw it on one of the YT videos. Can't remember the camera though.

I might also be full of it, and it might not be what you're looking for.
As far as I can tell, the pre-MF feature on Olympus cameras sets the focus to a pre-determined distance. I'm looking for a method of taking a sequence of 3 shots - one at current focus position, and one with focus shifted slightly forwards and backwards from the current focus position - ie. focus bracketing, in very fine steps.
 
I think your best course of action is to PM Jim Kasson -- because:

1) current Nikon step sizes might be smaller than on Z7(i) that he tested. This is possible because a) Nikon has released multiple full-frame camera models and firmware updates since that Z7, and b) I think (but am not sure, I could easily be confusing with a different brand) they've increased the number of possible steps in a bracketing run, which might imply a reduction in step size.

2) I believe he continues to use Nikon and Fuji GFX, and therefore might have had further interest and researched other models of those brands -- but not bothered to publish results, (perhaps because he was seriously injured in a fall a few years back).

If you do contact him, perhaps summarize plus provide a link to your original lengthy post here, rather than fully restate it, because the PM system here at DPR used to be relatively-limited in length allowed. Not sure if that's still true since DPR changed hands.

Another possibility might be to repost your original query (modified as required based on your subsequent reading of Kasson's focus-bracketing step-size post), over at the "Photographic Science and Technology" forum here at DPR.
 
I think your best course of action is to PM Jim Kasson -- because:

1) current Nikon step sizes might be smaller than on Z7(i) that he tested. This is possible because a) Nikon has released multiple full-frame camera models and firmware updates since that Z7, and b) I think (but am not sure, I could easily be confusing with a different brand) they've increased the number of possible steps in a bracketing run, which might imply a reduction in step size.

2) I believe he continues to use Nikon and Fuji GFX, and therefore might have had further interest and researched other models of those brands -- but not bothered to publish results, (perhaps because he was seriously injured in a fall a few years back).

If you do contact him, perhaps summarize plus provide a link to your original lengthy post here, rather than fully restate it, because the PM system here at DPR used to be relatively-limited in length allowed. Not sure if that's still true since DPR changed hands.

Another possibility might be to repost your original query (modified as required based on your subsequent reading of Kasson's focus-bracketing step-size post), over at the "Photographic Science and Technology" forum here at DPR.
Thanks for your advice - all good points. I will consider acting on them. I had already considered posting something on the Photographic Science and Technology forum.

TBH I'm surprised how few people seem to be interested in this method of working themselves. There are endless posts about AF focus accuracy but very few about trying to bracket to ensure a good result. I thought I'd get some responses from people who use (or would like to use) the method.
 
I think your best course of action is to PM Jim Kasson -- because:

1) current Nikon step sizes might be smaller than on Z7(i) that he tested. This is possible because a) Nikon has released multiple full-frame camera models and firmware updates since that Z7, and b) I think (but am not sure, I could easily be confusing with a different brand) they've increased the number of possible steps in a bracketing run, which might imply a reduction in step size.

2) I believe he continues to use Nikon and Fuji GFX, and therefore might have had further interest and researched other models of those brands -- but not bothered to publish results, (perhaps because he was seriously injured in a fall a few years back).

If you do contact him, perhaps summarize plus provide a link to your original lengthy post here, rather than fully restate it, because the PM system here at DPR used to be relatively-limited in length allowed. Not sure if that's still true since DPR changed hands.

Another possibility might be to repost your original query (modified as required based on your subsequent reading of Kasson's focus-bracketing step-size post), over at the "Photographic Science and Technology" forum here at DPR.
Thanks for your advice - all good points. I will consider acting on them. I had already considered posting something on the Photographic Science and Technology forum.

TBH I'm surprised how few people seem to be interested in this method of working themselves. There are endless posts about AF focus accuracy but very few about trying to bracket to ensure a good result. I thought I'd get some responses from people who use (or would like to use) the method.
I hope you can find a way. FWIW, IIRC, I think early models of the Canon EOS (EF-mount) cameras, still in film days, use to operate the way you want. (EDIT: That is, 3 shots, 0 + -, no idea about how big the step). (Wasn't a Canon user then, so I might be wrong about that).
 
Last edited:
I think your best course of action is to PM Jim Kasson -- because:

1) current Nikon step sizes might be smaller than on Z7(i) that he tested. This is possible because a) Nikon has released multiple full-frame camera models and firmware updates since that Z7, and b) I think (but am not sure, I could easily be confusing with a different brand) they've increased the number of possible steps in a bracketing run, which might imply a reduction in step size.

2) I believe he continues to use Nikon and Fuji GFX, and therefore might have had further interest and researched other models of those brands -- but not bothered to publish results, (perhaps because he was seriously injured in a fall a few years back).

If you do contact him, perhaps summarize plus provide a link to your original lengthy post here, rather than fully restate it, because the PM system here at DPR used to be relatively-limited in length allowed. Not sure if that's still true since DPR changed hands.

Another possibility might be to repost your original query (modified as required based on your subsequent reading of Kasson's focus-bracketing step-size post), over at the "Photographic Science and Technology" forum here at DPR.
Thanks for your advice - all good points. I will consider acting on them. I had already considered posting something on the Photographic Science and Technology forum.

TBH I'm surprised how few people seem to be interested in this method of working themselves. There are endless posts about AF focus accuracy but very few about trying to bracket to ensure a good result. I thought I'd get some responses from people who use (or would like to use) the method.
Autofocus is actually very accurate for stationary subjects, particularly if the selected focus point has significant sharp contrast. Just use a small box and put it over the focus point.

If I had a subject with significant depth my focus point would typically be 1/3 to 1/2 of the way from front to back, Any of the DOF calculators will give you the actual front/back distances.

If the subject depth exceeded the DOF, then I'd stack.

Some cameras will let you preview at the selected aperture, which further decreases the need for multiple images focused at different distances.

--

Sherm

Sherms flickr page

P950 album

P900 album RX10iv album
OM1.2 150-600 album
 
I think your best course of action is to PM Jim Kasson -- because:

1) current Nikon step sizes might be smaller than on Z7(i) that he tested. This is possible because a) Nikon has released multiple full-frame camera models and firmware updates since that Z7, and b) I think (but am not sure, I could easily be confusing with a different brand) they've increased the number of possible steps in a bracketing run, which might imply a reduction in step size.

2) I believe he continues to use Nikon and Fuji GFX, and therefore might have had further interest and researched other models of those brands -- but not bothered to publish results, (perhaps because he was seriously injured in a fall a few years back).

If you do contact him, perhaps summarize plus provide a link to your original lengthy post here, rather than fully restate it, because the PM system here at DPR used to be relatively-limited in length allowed. Not sure if that's still true since DPR changed hands.

Another possibility might be to repost your original query (modified as required based on your subsequent reading of Kasson's focus-bracketing step-size post), over at the "Photographic Science and Technology" forum here at DPR.
Thanks for your advice - all good points. I will consider acting on them. I had already considered posting something on the Photographic Science and Technology forum.

TBH I'm surprised how few people seem to be interested in this method of working themselves. There are endless posts about AF focus accuracy but very few about trying to bracket to ensure a good result. I thought I'd get some responses from people who use (or would like to use) the method.
Autofocus is actually very accurate for stationary subjects, particularly if the selected focus point has significant sharp contrast. Just use a small box and put it over the focus point.
That's the generally accepted wisdom, but in practice, I'd say it's not all that true. With the point over an easy area of detail, cameras will usually get this in focus, but especially with zoom lenses with smaller max apertures, there is a range of focusing distances where this area will remain 'in focus', but where it is set within this range can affect objects at the edge of the DOF quite a bit. In a recent test on a tripod, using AF on exactly the same point, then defocusing and re-focusing, I found focus distance was set differently between shots, which was quite apparent in objects at the periphery of the DOF. There was an even bigger difference when switching lenses.

In addition to this, subjects in practice are rarely ideal for AF - receding planes, low contrast, foliage with depth, atmospheric haze etc, and quite often there isn't anything to focus on at the required distance.
If I had a subject with significant depth my focus point would typically be 1/3 to 1/2 of the way from front to back, Any of the DOF calculators will give you the actual front/back distances.
DOF calculations are theoretical and not particularly accurate in the field since they will rely on estimations of distance. They will be no better than my best judgement, having done this for many years, and slow down the process immensely. They also have no idea about field curvature, which in a complex zoom lens is almost impossible to predict. In fact, this is one of the major reasons why I would like to have a workable focus bracketing feature.
If the subject depth exceeded the DOF, then I'd stack.

Some cameras will let you preview at the selected aperture, which further decreases the need for multiple images focused at different distances.
Not really. The resolution of most EVFs simply isn't sufficient to see what I need to see at normal magnification, and in magnified view they generally become to noisy when stopped down, not to mention that it would take too long to check enough parts of the image like this. I've tried all these techniques and nothing really does the job - that's why I'm asking about bracketing.
 

Keyboard shortcuts

Back
Top