Very true - people forget that ALL of this stuff is in its infancy. In fact, we're thinking of selling our A7R4 because of the issue with the 200-600.
Subject tracking with Nikon Z6II firmware 1.2
- Thread starter Rassie
- Start date
If you would like to post, you'll need to register. Note that if you have a BCG store account, you'll need a new, separate account here (we keep the two sites separate for security purposes).
One thing I'll add to the tracking discussion is the key is to have initial focus on the subject. I've shot bursts that have one frame sharp out of a series, but to me it's obvious that the one sharp frame is because the subject flew into the focus plane, not because I ever acquired focus. AF tracking is not a strategy where you hope the camera picks up a moving subject.
The issue is about subject tracking and not getting that right in Z ii bodies is a serious design & implementation issue .
One need to note that Sony's A1 software was built over what was implemented in A9 series .Nikon to get in right in Z 9 has to get it right in Z ii bodies ( so that the software developers know whether the algorithm works in real world situations or not) since there has been no other announcement about Z 8 before Z9 & one can assume now that Z9 is the next body from Z9 will all bells & whistles.
Agree 100%. The issue I see with the Z cameras is that they are often reluctant to lock onto an already moving target. Initial locks are a big issue and if you don't have it, there's no way the camera will track (although, even when the camera does track a faster moving target, it often lets go of the target after a moment and hits the background instead).
abc123brian
Well-known member
The point about readout times and processing power is still valid when it comes to the algorithm. If the sensor readout time is a limiting factor, which we know it is, the algorithm has to be written with that limitation in mind. While they may reuse the old algorithm and it will just work faster in the Z9, the possibilities of faster readout times and more processing power means entirely new things are possible and a completely different algorithm could be used.
Here's the video I promised 
New algorithm would require design changes, extensive testing & further delays which would only help the competition.That would be the last option when Nikon can not get it right by tweaking the exisitng algorthim beyond a deadline
The sensor readout is the limiting factor for EVF performance. There is also an impact on AF speed and rolling shutter. Nikon has already communicated features indicating that readout with the Z9 is going to be at a typical flagship mirrorless rate of around 120 Hz like the Sony A1, and that 4 times the rate of the Z6II/Z7II. That's twice the speed of the A9II. The Canon R3 will be in the same range. That's the game for a $6500 flagship level camera.
They don't need to start over. The readout speed is a known issue with a known solution. But the solution is integrated - it needs to align camera lenses for a flagship body, sensor and sensor software, updated hardware for a number of components, and of course typical firmware and upgrades. Put together, there are a lot of moving pieces and things that need to be done. Much of it is leveraging D6 and Z6II/Z7II cameras and updates, but it's more than that.
The central strategy Nikon has been using for years is creating incremental advances and then reusing them across multiple cameras.
DaveR
Active member
Wow. What a cool thread. I've learnt a lot following it. Let's see if I've got this right. The key to the performance of the Z9 is the new stacked sensor that Nikon have designed. The limitations of the current Z series cameras is down to using sensors from their current dslr's. Add in the new Z series lenses to the mix which should further increase performance along with more processing power and better software/firmware and the Z9 is looking like an exciting prospect. I do hope they come up with something special with the new Z series 600 f4. That would be the icing on the cake.
I think Nikon has a real problem.
I decided to switch to mirrorless this winter (saved the money during lockdown) and ultimately decided to stay with Nikon. I shoot events, street, journalism, performance (bands, acts and video productions), landscape and environmental portraits, not much BIF, so my use cases are probably different from most here.
In the past I've rented a Z7 because of its silent mode for BTS shooting on a video production set, so I knew the ii's would be a least that good. I have a couple of Sony mirrorless and have gotten terrific pictures from them so I really considered buying into their systems. I think the deciding factor for Nikon was the stream of reports on the quality of the new Z glass, and also my experience with Nikon ergonomics and reliability over the past decades.
I duplicated my go-to, holy-trinity, gold ring F lenses with Z S-lenses, and bought Z6ii and Z7ii bodies. Got an FTZ adapter so I could use my macro's until Nikon duplicates these in Z-land.
On net, this has been an useful upgrade...The IBIS, WYSIWYG, tilting, hi-rez screen, playback through the EVF, silent shutter, focus stacking, razor sharp lenses.
But the autofocus has been a major disappointment. I find myself shooting in single point, or group focusing just like on a DSLR. I shoot in variable, low, and backlight using shallow depth-of-field to separate out subjects from messy backgrounds (just like shooting birds). The fancy auto, intelligent face/eye/tracking just can't be trusted. Of course, I can put the focus anywhere and that is an upgrade from DSLR. But its is slow, and clumsy, to be desperately toggling the joy stick trying to keep up with a subject when the composition needs to change. This has been a real disappointment, particularly when Sony and Canon apparently have automated autofocus nailed.
I haven't completely retired the DSLR bodies, either, since my macro's are screwdriver focus lenses. Also, the built-in flash on my D810 can be used to trigger remote flashes without bothering with separate wireless transceivers.
If this was a completely analytical decision, unfettered by decades of great experience with Nikon, I probably would have (in retrospect) purchased Canon or Sony. Ultimately, switching to mirrorless, one is buying into a completely new system so from a cost/disruption viewpoint all the manufacturers start out on an (roughly) even footing for my use cases. There is still a learning curve moving to mirrorless irrespective of brand. When I dip into my legacy Nikkor lenses, I am using legacy bodies as well...I haven't actually used the FTZ adapter at all during this transition (to my surprise). The mirrorless side could have been from any manufacturer with no additional disruption.
I think Nikon is very vulnerable. They just narrowly missed losing me as a customer. Buying Zii's, Z glass and accessories this past winter has been like searching for diamonds on a muddy road. Their Z-lens roadmap is far from complete while the competition, particularly Sony, rolls out competitively-performing lenses with more advanced designs resulting in smaller sizes and lower weights.
Nikon mirrorless investors are left pretty much waiting for the next shoe to fall...For better or worse. Pundits talk of the Z9 as if it were a real camera...Surely a sign of desperation. Given the shortfalls of the Z6/7's relative to the competition, to expect a Z9 to appear matching or exceeding their D6, D850 or D500 DSLR's, and also the performance of their top mirrorless competitors, is more hope than certainty.
Nikon has pulled out miracles in the past. Let's hope they do it again.
I decided to switch to mirrorless this winter (saved the money during lockdown) and ultimately decided to stay with Nikon. I shoot events, street, journalism, performance (bands, acts and video productions), landscape and environmental portraits, not much BIF, so my use cases are probably different from most here.
In the past I've rented a Z7 because of its silent mode for BTS shooting on a video production set, so I knew the ii's would be a least that good. I have a couple of Sony mirrorless and have gotten terrific pictures from them so I really considered buying into their systems. I think the deciding factor for Nikon was the stream of reports on the quality of the new Z glass, and also my experience with Nikon ergonomics and reliability over the past decades.
I duplicated my go-to, holy-trinity, gold ring F lenses with Z S-lenses, and bought Z6ii and Z7ii bodies. Got an FTZ adapter so I could use my macro's until Nikon duplicates these in Z-land.
On net, this has been an useful upgrade...The IBIS, WYSIWYG, tilting, hi-rez screen, playback through the EVF, silent shutter, focus stacking, razor sharp lenses.
But the autofocus has been a major disappointment. I find myself shooting in single point, or group focusing just like on a DSLR. I shoot in variable, low, and backlight using shallow depth-of-field to separate out subjects from messy backgrounds (just like shooting birds). The fancy auto, intelligent face/eye/tracking just can't be trusted. Of course, I can put the focus anywhere and that is an upgrade from DSLR. But its is slow, and clumsy, to be desperately toggling the joy stick trying to keep up with a subject when the composition needs to change. This has been a real disappointment, particularly when Sony and Canon apparently have automated autofocus nailed.
I haven't completely retired the DSLR bodies, either, since my macro's are screwdriver focus lenses. Also, the built-in flash on my D810 can be used to trigger remote flashes without bothering with separate wireless transceivers.
If this was a completely analytical decision, unfettered by decades of great experience with Nikon, I probably would have (in retrospect) purchased Canon or Sony. Ultimately, switching to mirrorless, one is buying into a completely new system so from a cost/disruption viewpoint all the manufacturers start out on an (roughly) even footing for my use cases. There is still a learning curve moving to mirrorless irrespective of brand. When I dip into my legacy Nikkor lenses, I am using legacy bodies as well...I haven't actually used the FTZ adapter at all during this transition (to my surprise). The mirrorless side could have been from any manufacturer with no additional disruption.
I think Nikon is very vulnerable. They just narrowly missed losing me as a customer. Buying Zii's, Z glass and accessories this past winter has been like searching for diamonds on a muddy road. Their Z-lens roadmap is far from complete while the competition, particularly Sony, rolls out competitively-performing lenses with more advanced designs resulting in smaller sizes and lower weights.
Nikon mirrorless investors are left pretty much waiting for the next shoe to fall...For better or worse. Pundits talk of the Z9 as if it were a real camera...Surely a sign of desperation. Given the shortfalls of the Z6/7's relative to the competition, to expect a Z9 to appear matching or exceeding their D6, D850 or D500 DSLR's, and also the performance of their top mirrorless competitors, is more hope than certainty.
Nikon has pulled out miracles in the past. Let's hope they do it again.
Kenny88
Active member
Hi Steve: Thank you for the video highlighting the 1.2.0 firmware update - appreciate the time that you took with regards.
I don't own any of the Z cameras yet and have been sitting on the sidelines waiting to hear when and if Nikon catches up with Canon or Sony on the tracking and focusing front. I'm getting tired of waiting, and there's a chance that I'm going to bail out from Nikon - sad because I want to stick with the brand, but I'm sure there's others in the same boat as me - a shame that they can't make any headway in keeping up with the competition.
With the rumored A7IV coming up, which might adopt the same AF and tracking as the Sony A1, Nikon is going to sink faster.
Take care
I don't own any of the Z cameras yet and have been sitting on the sidelines waiting to hear when and if Nikon catches up with Canon or Sony on the tracking and focusing front. I'm getting tired of waiting, and there's a chance that I'm going to bail out from Nikon - sad because I want to stick with the brand, but I'm sure there's others in the same boat as me - a shame that they can't make any headway in keeping up with the competition.
With the rumored A7IV coming up, which might adopt the same AF and tracking as the Sony A1, Nikon is going to sink faster.
Take care
Demodave
Active member
Ive shot the Z7II and sony A9ii side by side for this last week. The A9II w and wo the 1.4 and the 200-600. The Z7II with the 500PF and the 1.4. The Sony is quicker and more responsive for BIF hands down. In fact no comparison. The further away the subject was the closer the z7II was to the sony. The closer the subject was, The better the Sony did. For sitters or birds on a stick, the Z7II may even have an edge on the Sony. But my opinion is the 500PF is sharper at a distance than the 200-600. The 200-600 is however sharp. I just give the slight edge to Nikon
For quick movers, ie Osprey and Pelicans diving, I got to the point I wouldn't even pick up the Z7II. Absolutely no comparison.
For panning shots, you get all shots in focus and sharp with the A9II. I was getting missed focus with the Z.
I'm a Nikon guy from the early 80's. Right now,for this trip. I'm very glad I have the sony. In fact I'm considering moving to all Sony and picking up the 600 G master.
The ONLY thing holding me back is the 500PF.
The Z's and Nikon better get their act in gear quickly with a 2.0 firmware AF fix or I'm finding myself in the odd position of parting with Nikon. And thats hard to do as an NPS member.....
For quick movers, ie Osprey and Pelicans diving, I got to the point I wouldn't even pick up the Z7II. Absolutely no comparison.
For panning shots, you get all shots in focus and sharp with the A9II. I was getting missed focus with the Z.
I'm a Nikon guy from the early 80's. Right now,for this trip. I'm very glad I have the sony. In fact I'm considering moving to all Sony and picking up the 600 G master.
The ONLY thing holding me back is the 500PF.
The Z's and Nikon better get their act in gear quickly with a 2.0 firmware AF fix or I'm finding myself in the odd position of parting with Nikon. And thats hard to do as an NPS member.....
smkunder
Active member
I think there are a lot of us that have similar feelings.
wild-pics
Member
Agree 100%. However checking the retail price just now here in Australia the Z6ii is $3400 aud and A9ii is 6600 aud. I doubt that the Zi/ii's were ever designed for action so I only use my Z7 for travel. I think we'll have to wait for different Z bodies that are more suited for action to be released. For the time being I will use the D850 and 500PF for wildlife.
Demodave
Active member
I picked up the Z6II for 2K. The Z7II for 3K. And a mint A9II with 372 shutter count for 3.5K
Cost wise per dollar. The A9II outperforms either of the Z's
Cost wise per dollar. The A9II outperforms either of the Z's
Demodave
Active member
I'm just now finishing a vaca down in FL. One of the thougjts was to leave the DSLRs at home and just compare and really test out the mirrorless on birds. While I have over 5K worth of pics to analyze and post process after 5 days of basically non stop shooting, my initial thoughts are as I posted above. I'm not comparing model for model. I'm running system against system based on what's available.
If I were just shooting mammals. If pick up a 600E from Nikon and trade in my 600G. Id run my 500Pf and 300PF. And 70-200E and be happy.
I am enjoying the mirrorless benefits. And fully imtend to transition to all mirrorless. For me, I enjoy it more.
Sony's 600 F4 is arguably sharper than Nikons, especially wIth the1.4 and 2x TC.
That is part of the driving force.
In the end for me it's about weight, IQ, and most importantly results
This week on birds. My results were better with the Sony mirrorless than with the Nikon Z's.
As to Z9 vs A1/A9? The Sony is in the market. The Nikon is a hope. In the mean time. I have pics on my computer.......
If I were just shooting mammals. If pick up a 600E from Nikon and trade in my 600G. Id run my 500Pf and 300PF. And 70-200E and be happy.
I am enjoying the mirrorless benefits. And fully imtend to transition to all mirrorless. For me, I enjoy it more.
Sony's 600 F4 is arguably sharper than Nikons, especially wIth the1.4 and 2x TC.
That is part of the driving force.
In the end for me it's about weight, IQ, and most importantly results
This week on birds. My results were better with the Sony mirrorless than with the Nikon Z's.
As to Z9 vs A1/A9? The Sony is in the market. The Nikon is a hope. In the mean time. I have pics on my computer.......
Last edited:
fcotterill
Well-known member
Well said @Steve, Maturity is indeed the crux of the issues. Assuming Nikon meets key targets on the Z lenses Roadmap, native Z mount options will be quite different by December. This assumes there are thousands prepared to pay premium prices for exotic Z telephotos. The 70-200 f2.8S, 100-400 S, the pair of ZTCs and 200-600 Z will probably sell well, compared to 400 f2.8S and 600 f4S. The sales of F-exotics at huge costs suggest the majority of Nikon shooters will continue to rely on the FTZ for E and G telephotos.
IMHO +IME, as far as the complete lens lineup will soon stand, the 100-400 S [my emphasis added ie excellent IQ on MILC AF] is the most significant addition/upgrade. It will surely improve on the jaded 80-400 G, especially if it maintains decent AF and IQ with the ZTC14.
Even so if one compares "Lineup-Completeness" of current ILC Systems to meet key demands of respective genres... What is the key issue? for Nikon's Greater Ecosystem, the current gap has been (is being) discussed ad nauseum - as yet no Z camera to match the D5/D6 for active subjects. Otherwise, the reality is obvious. We have never enjoyed the vast choice of so many excellent F Nikkors and F-fit 3rd party lenses to meet the needs of all genres. The reality considering all the lenses, it's Money that is the persisting constraint for almost all of us!
In the meantime, the serious Nikon shooter is constrained only where they must have a Mirrorless-only system that delivers BOTH D5/D6 action+Pro-level haptics with silent-shooting (given F telephotos deliver cutting edge performance with the FTZ). Until the Z9 is post-launch and actually on shelves, for wildlife, sports, studio action, when is a D6 (or D850) the critical showstopper? How many of us have to have an action MILC with silent-shooting+EVF to get keepers/pay bills?
Otherwise, use a Z II for silent shooting (equally travel, events, landscapes etc). The D5/D6 are still the industry leading cameras for action, with a D500 / D850 almost as excellent and more affordable options.
Actually, of the DSLRs I find the D780 is bigger gap and disappointment, seriously underwhelming haptics..... To conclude, there's a big Pachyderm-of-a-Gap in the room wrt Nikon's current and anticipated cameras. Few question how the Z9 with 8K (ie high resolution FX sensor) yet beating the D6 AF etc etc etc can also deliver D6 level IQ at high ISOs (?) Considering the underwhelming Z6 and D780, Nikon has yet to fill the niche with a prosumer ILC combining excellent lowlight IQ and D5 level AF and Haptics.
IMHO +IME, as far as the complete lens lineup will soon stand, the 100-400 S [my emphasis added ie excellent IQ on MILC AF] is the most significant addition/upgrade. It will surely improve on the jaded 80-400 G, especially if it maintains decent AF and IQ with the ZTC14.
Even so if one compares "Lineup-Completeness" of current ILC Systems to meet key demands of respective genres... What is the key issue? for Nikon's Greater Ecosystem, the current gap has been (is being) discussed ad nauseum - as yet no Z camera to match the D5/D6 for active subjects. Otherwise, the reality is obvious. We have never enjoyed the vast choice of so many excellent F Nikkors and F-fit 3rd party lenses to meet the needs of all genres. The reality considering all the lenses, it's Money that is the persisting constraint for almost all of us!
In the meantime, the serious Nikon shooter is constrained only where they must have a Mirrorless-only system that delivers BOTH D5/D6 action+Pro-level haptics with silent-shooting (given F telephotos deliver cutting edge performance with the FTZ). Until the Z9 is post-launch and actually on shelves, for wildlife, sports, studio action, when is a D6 (or D850) the critical showstopper? How many of us have to have an action MILC with silent-shooting+EVF to get keepers/pay bills?
Otherwise, use a Z II for silent shooting (equally travel, events, landscapes etc). The D5/D6 are still the industry leading cameras for action, with a D500 / D850 almost as excellent and more affordable options.
Actually, of the DSLRs I find the D780 is bigger gap and disappointment, seriously underwhelming haptics..... To conclude, there's a big Pachyderm-of-a-Gap in the room wrt Nikon's current and anticipated cameras. Few question how the Z9 with 8K (ie high resolution FX sensor) yet beating the D6 AF etc etc etc can also deliver D6 level IQ at high ISOs (?) Considering the underwhelming Z6 and D780, Nikon has yet to fill the niche with a prosumer ILC combining excellent lowlight IQ and D5 level AF and Haptics.
Last edited:
fcotterill
Well-known member
A fundamental dichotomy challenges the strategists and engineer teams designing a line up of mirrorless ILCs, compared to what paid off previously rather well in the R&D strategy that put out updated models of DSLRs. Prior to the Z system, Nikon must have earned decent returns by packaging the D4 sensor into the Df, cloning the D5 AF engine into the D850 and D500; maintaining the D5 sensor in the D6 etc.
Cloning across core units and parts is still possible in some aspects of MILCs. The sensor is the exception, given it packs in so much of the AF. This injects a challenge to follow up a bleeding edge ILC - in the shape of a Z9 - with more affordable models in a similar niche. Presumably, this challenge is magnified in action cameras built on a stacked-sensor. We read stacked sensors not only require long lead-ins of R&D, but they are challenging and costly to fabricate.
Equally, as argued above, it's hard to see how further upgrades of software and processors can rectify performance deficits in the Z6 and Z7 sensors. The inherent design of this sensor architecture appears to constrain the AF and frame rates, which yet again limits the feasibility of engineers to maintain one particular sensor as the core unit for a series of upgraded cameras. The corollary is Nikon will (are) probably be designing new sensors for the Z III series.
This constraint must have been stressing out Nikon's engineers for some time, assuming R&D for the Z system (likely Z9 also) started a few years prior to 2018! Perhaps, their plan to solve the dilemma of a prosumer Z camera for action genres will revisit the DX style strategy. Scale down the Z9 stacked-sensor into a Z90 - assuming (1) the sensor is the major cost of all the components, and (2) fabrication yields and costs are proportionally less for smaller sensors.
Just a hypothesis
Cloning across core units and parts is still possible in some aspects of MILCs. The sensor is the exception, given it packs in so much of the AF. This injects a challenge to follow up a bleeding edge ILC - in the shape of a Z9 - with more affordable models in a similar niche. Presumably, this challenge is magnified in action cameras built on a stacked-sensor. We read stacked sensors not only require long lead-ins of R&D, but they are challenging and costly to fabricate.
Equally, as argued above, it's hard to see how further upgrades of software and processors can rectify performance deficits in the Z6 and Z7 sensors. The inherent design of this sensor architecture appears to constrain the AF and frame rates, which yet again limits the feasibility of engineers to maintain one particular sensor as the core unit for a series of upgraded cameras. The corollary is Nikon will (are) probably be designing new sensors for the Z III series.
This constraint must have been stressing out Nikon's engineers for some time, assuming R&D for the Z system (likely Z9 also) started a few years prior to 2018! Perhaps, their plan to solve the dilemma of a prosumer Z camera for action genres will revisit the DX style strategy. Scale down the Z9 stacked-sensor into a Z90 - assuming (1) the sensor is the major cost of all the components, and (2) fabrication yields and costs are proportionally less for smaller sensors.
Just a hypothesis
Eddietheone
Member
I Think Rassie instead of praiseing the camera it was proberly your skill with panning the camera that made the difference. Can you teach us all how to do it please
Thanks for that, but I think I was just lucky at that specific time. I was near the airport yesterday where I again experimented with subject tracking on a red-tailed hawk as well as aircraft coming in for landing. The sky was partially clouded over. Suffice to say that the camera's tracking performance sucked, it kept losing tracking and focus on the subject. My thoughts at this time are that the tracking speed of the tracking box was improved with the latest firmware update in that when it's locked on it responds better in sticking with an erratic subject. With the previous firmware in the past the box easily drifted off the subject when it made sudden course changes.
What became clear to me yesterday, and as borne out in Steve's testing videos, if the background is not a nice, even blue sky, anything else that shows in the frame is likely going to pull the tracking box off the subject. This goes for any background foliage or terrain, or even clouds with some definition in them. So what I think at this time is that I'll consider using the subject tracking on a bird flying in clear sky with nothing else in the frame. For anything else Wide Small or Wide Large are better options to track moving subjects.
I firmly believe Nikon will get this right with future firmware. It's just a matter of time. I have too much time, effort and money invested in Nikon gear to consider jumping ship to another brand at this late stage of my life, so I'm somewhat impatiently waiting for Nikon to fix things. Thank heavens that BIF photography is only a part of what I normally shoot so for everything else I do the camera is great. Having said that I see myself doing more and more raptor hunting photography going forward. Late last summer I found a great spot at the west end of Lake Ontario where there is quite a lot of action with terns, ospreys, kingfishers and bald eagles diving for fish so I'll be spending a lot of time there going forward. And then it will be nice to have subject tracking that works as intended.
Last edited:
ccirelli
Well-known member
@Rassie thanks for this thread, good discussion & observations. Have you experimented with Auto Area AF for BIF? I've heard some reports of much greater success in the mark II models. My hunch is the results may be similar if the background isn't clean, but it's worth a try.
I don't use Subject Tracking for faster action (and I realize you're just running performance tests here too), I'm in Wide Small or Dynamic Area (is Dynamic Area even still needed?) or even Auto Area. But I will say for almost everything else lately - certainly portraits - Subject Tracking / eye detection is just phenomenal. We don't have a wildlife / bird mode yet, only "dogs & cats" (lol) and if they have dark fur, it's not reliable (try eye-detecting a black & white Border Collie).
I'm hopeful Nikon takes a few notes from Canon's wildlife / bird detection AF. I'm not a wildlife-primary shooter, so the current Z's are meeting all my needs - but it would be great to have some equivalent AF programming (it can be done).
I don't use Subject Tracking for faster action (and I realize you're just running performance tests here too), I'm in Wide Small or Dynamic Area (is Dynamic Area even still needed?) or even Auto Area. But I will say for almost everything else lately - certainly portraits - Subject Tracking / eye detection is just phenomenal. We don't have a wildlife / bird mode yet, only "dogs & cats" (lol) and if they have dark fur, it's not reliable (try eye-detecting a black & white Border Collie
).I'm hopeful Nikon takes a few notes from Canon's wildlife / bird detection AF. I'm not a wildlife-primary shooter, so the current Z's are meeting all my needs - but it would be great to have some equivalent AF programming (it can be done).
I have not tried Auto-area AF yet on BIF. I guess I've always been worried about losing too much control for focus points. For BIF currently I use mostly Wide Area Small and Large.
abc123brian
Well-known member
Just a note from a recent outing, I spent some time shooting a black bear in an open field with my Z7ii recently. I switched to Auto Area AF to see how the camera would perform. When the bear looked up, it found the eyes and tracked them. When he turned his head some, it would switch to face detect. I certainly wouldn’t depend on this mode, as when the bear looked down, it went back to auto area and focuses on the bear’s back. I honestly didn’t expect it to work at all on the bear since they are difficult already, but I did find it potentially promising for future firmwares.
I've had a similar experience with people. If you are photographing a line of people from the side and one person looks away (now in profile), then there is no longer an eye box, or even a face box, and there's no way to pick them up again without turning off eye focus. Happens in the case of photographing musicians in a line on stage. You can lose a decisive moment when this happen. Cartier-Bresson never had to deal with this!