Z9 1st Impressions & Firmware Wishlist: Brad Hill

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).

Member's gallery
This shoutout for 3D tracking in the D6 by Jeffrey Wu only came to my attention yesterday; this prompted by his report on PP Z9 images of high ISOs... This is a repost of his D6 promo video. Previously I was only aware of Moose Peterson reporting on the the major improvement in D6 subject tracking - using AI to focus on grouse and sheep [correction, Moose was endorsing Auto-AF mode in the D6 using Custom GroupAF with Eye-detection turned on, and also AutoAF]

I've singled this out here because 1. Nikon presumably used the underlying AI algorithm from the Z9, and 2. Nikon can be relied on be infallible about failing to spell out the benefits and intricacies of what-under-the-hood, in their cameras especially!

Underreporting and Under-explaining how their flagship autofocus works defines a thread in the DNA of this infuriating company!!!!
 
Last edited:
Member's gallery
arbitrage said:
An excellent article. I think Brad only missed one thing in it and that is the ability of Wide Area AF to continue tracking a detected subject outside of the Wide Area box (with the box not on the subject anymore at all) for a given amount of time determined by the Blocked Shot Response setting (5 being the longest amount of time before it tries to refocus under the Wide Area).
Click to expand...
Does Wide Area track at all? I thought only 3D does tracking.
 
Member's gallery
This topic seems to be the most confusing as there are lot of conflicting information in different manuals/guides.

According to brad hill, blocked shot response has no effect on any of the AF modes except dynamic area AF, IF subject detection is ON. How it works when subject detection is off is anyone's guess for now. There's conflicting information between Nikon's reference guide vs AF settings guide for sports.

John Woodworth said:
Blocked Shot AF Response functions in Mode 3 for 3D Tracking and Modes 3,4,5 for Auto Area. See p600 of Reference Guide Z9RG_02.
Click to expand...
 
Member's gallery
thelordofthelight said:
This topic seems to be the most confusing as there are lot of conflicting information in different manuals/guides.

According to brad hill, blocked shot response has no effect on any of the AF modes except dynamic area AF, IF subject detection is ON. How it works when subject detection is off is anyone's guess for now. There's conflicting information between Nikon's reference guide vs AF settings guide for sports.
Click to expand...

And yet my testing very clearly showed that the Blocked shot response is functioning in Wide Area with Subject Detect ON because it would continue to track the subject significantly longer when the subject completely left the defined Wide Area at a setting of 5 than it would with a setting of 1. Simple to test if you have a Z9. Start wide Area AF tracking a test subject (say stuffed animal) and then move the Wide Area way off the animal over a distant background while continuing to hold down AF. Do this a few times going back and forth between 1 and 5. You will clearly notice a time difference before the AF tries to refocus on whatever is under the Wide Area AF box/rectangle.
 
Member's gallery
Barbara said:
So is 3D tracking also subject to the blocked shot time delay?
Click to expand...
After reading John's reply I will qualify my reply....Blocked shot should affect 3D but only if the system can't recognize the subject anymore. Otherwise it will continue to track forever if the subject is somewhere in the entire frame. In Wide Area it has more of an affect because you have to keep the Wide Area on the subject and the setting for Blocked Shot affects how soon you have to get back on your subject before it tries to refocus under the defined Wide Area box/rectangle.

But then we have confusion of Brad Hill saying it doesn't affect 3D, Auto or Wide-Area if Subject detect is ON. I don't believe this to be true as it clearly affects Wide-Area with subject detect ON.
 
Member's gallery
arbitrage said:
After reading John's reply I will qualify my reply....Blocked shot should affect 3D but only if the system can't recognize the subject anymore. Otherwise it will continue to track forever if the subject is somewhere in the entire frame. In Wide Area it has more of an affect because you have to keep the Wide Area on the subject and the setting for Blocked Shot affects how soon you have to get back on your subject before it tries to refocus under the defined Wide Area box/rectangle.

But then we have confusion of Brad Hill saying it doesn't affect 3D, Auto or Wide-Area if Subject detect is ON. I don't believe this to be true as it clearly affects Wide-Area with subject detect ON.
Click to expand...
Have Nikon said that blocked shot response is different depending on whether subject detect is on or not? Some of these proclaimed gurus are less than accurate in their statements at times.
 
Member's gallery
Wes Peterson said:
Have Nikon said that blocked shot response is different depending on whether subject detect is on or not? Some of these proclaimed gurus are less than accurate in their statements at times.
Click to expand...
I don't think Nikon have said it anywhere official.

Check with fcotterill as he read it either on Brad Hill's blog or via private email with Brad Hill (I can't recall if Brad put this out publicly or not?)
 
Member's gallery
John Navitsky said:
Brad has discussed on his blog.

Natural Art Images: Voice: Brad Hill Blog

The Blog of Brad Hill, wildlife photographer. Cameras, lenses, camera gear, field tests, conservation, photo tours, and other bits of questionable wisdom from a young-at-heart nature photographer!
www.naturalart.ca www.naturalart.ca
Click to expand...
Except he didn't discuss anything about the Blocked Shot setting in that article. I think fcotterill got that info from private email with Brad.
 
Member's gallery
fcotterill said:
A useful overview explaining how to configure RSF, about which Nikon has provided even less explanation than is typical! Minor correction, in the D5 one can store only a single RSF setup across all 4 Custom Banks
Click to expand...
I think it would be far easier for the end users to follow the "Sony" approach or even the Z6 / Z7 approach where settings are saved to user modes. The "1, 2, 3) on the function dial. User 1, User 2 as was on my old D7200. That is a brilliant option. I set my camera for a given type of shooting, save the settings to User 1. Set it differently for a different scenario, save it to User 2, etc. If I make changes along the line and want to go back to a baseline, load User 1 and bingo I'm back. If I really like the changes I made save them to the user setting and boom there I have it.

This whole shooting banks and custom banks etc are confusing (at least to my feeble mind) and really do not behave in the manner as the User 1,2... settings. When I went from D7200 to D500 I saw the loss of those personalized settings as a step backward. I have configured buttons to handle most of my settings but still think it would be far easier to just give user configured options, save the settings and be done with it.

I do not shoot Sony (yet) but the 1, 2, 3 on the dial is a huge attraction to me where not having that capability with the Z9 is a dissatisfier in what appears to be an otherwise wonderful piece of technology.

OK, rant over sorry.
Jeff
 
Member's gallery
arbitrage said:
After reading John's reply I will qualify my reply....Blocked shot should affect 3D but only if the system can't recognize the subject anymore. Otherwise it will continue to track forever if the subject is somewhere in the entire frame. In Wide Area it has more of an affect because you have to keep the Wide Area on the subject and the setting for Blocked Shot affects how soon you have to get back on your subject before it tries to refocus under the defined Wide Area box/rectangle.

But then we have confusion of Brad Hill saying it doesn't affect 3D, Auto or Wide-Area if Subject detect is ON. I don't believe this to be true as it clearly affects Wide-Area with subject detect ON.
Click to expand...
I've tested this a good bit and in wide area mode if the camera locks on a subject it will track it outside the box indefinitely. The camera will also initially lock onto eyes that are not inside the focus area box if they are distinct enough. It behaves as if the focus box is more like a priority area for the camera to search first then it essentially goes into 3d mode. Perhaps is does re-test the area box on a time interval but it will clearly stay locked on to a strong signal outside the box indefinitely. This is easy enough for anyone to test in the comfort of your living room. If you don't have a willing subject dog or cat that will hold still long enough you can simply use prints of birds/wildlife.

It would be nice if Nikon would explain this stuff rather than us having to figure it out on our own. Great equipment, terrible company.
 
Last edited:
Member's gallery
NorthernFocus said:
I've tested this a good bit and in wide area mode if the camera locks on a subject it will track it outside the box indefinitely. The camera will also initially lock onto eyes that are not inside the focus area box if they are distinct enough. It behaves as if the focus box is more like a priority area for the camera to search first then it essentially goes into 3d mode. Perhaps is does re-test the area box on a time interval but it will clearly stay locked on to a strong signal outside the box indefinitely. This is easy enough for anyone to test in the comfort of your living room. If you don't have a willing subject dog or cat that will hold still long enough you can simply use prints of birds/wildlife.

It would be nice if Nikon would explain this stuff rather than us having to figure it out on our own. Great equipment, terrible company.
Click to expand...

This is actually pretty well explained in a video with Brad Hill and the Canadian Nikon guy (I'll try to find the link and post in a bit). The box defines where some part of the animal/person/subject needs to be. However if the eye/face is outside the box it will still go there and track it as long as the box is kept on some part of the animal. I've tested this on my cats in the house as well as various animals during my photo walks and it seems to bear out - as long as I have some part of the box touching the animal, it would find and focus on the eye or face. But if the animal left the box, it would try and acquire a target within the box.

I think the question still is how the blocked-shot settings affect things when there's an interruption/loss-of-detection of the subject - how quickly does it try and find something else vs re-acquire the original subject. I believe I read later that Brad (or someone else) determined that the settings have no effect on 3D or Auto, but it does affect the wide area modes, even if subject detection is on. That's slightly harder to test in a definitive way, but I think someone suggested a setup to test this and confirmed the settings do have an effect even with Subject detect on, and my fuzzy memory is also saying that Steve tried and reproduced the setup and confirmed the findings.

EDIT: The video is here:
and the discussion about the blocked-shot AF starts around 31min.
 
Last edited:
Member's gallery
NorthernFocus said:
I've tested this a good bit and in wide area mode if the camera locks on a subject it will track it outside the box indefinitely. The camera will also initially lock onto eyes that are not inside the focus area box if they are distinct enough. It behaves as if the focus box is more like a priority area for the camera to search first then it essentially goes into 3d mode. Perhaps is does re-test the area box on a time interval but it will clearly stay locked on to a strong signal outside the box indefinitely. This is easy enough for anyone to test in the comfort of your living room. If you don't have a willing subject dog or cat that will hold still long enough you can simply use prints of birds/wildlife.

It would be nice if Nikon would explain this stuff rather than us having to figure it out on our own. Great equipment, terrible company.
Click to expand...

That isn't the result I got with my testing. I used a pigeon on a rock just sleeping. Background 100s of feet away totally OOF. I got it doing the Eye AF on the pigeon and then moved the Wide Area way to the side over the background. Even on a setting of 5 it still did a refocus movement to try for the background after a second or two. On a setting of 1 it was very soon after moving it off the subject. If the Wide area stays on the subject, even a little bit, and it doesn't have to be on the head then it will hold the eye indefinetely. Probably even if the subject is just a little bit outside the marked box it will keep on it. But I repeatedly tested and moving the box aways from the bird caused the refocus drive while I continued to hold down AF-ON.

I wonder why we got different results??
 
Member's gallery
arbitrage said:
...I wonder why we got different results??
Click to expand...
NorthernFocus said:
Maybe distance to subject has something to do with it? All of my testing has been at close range with close BG. Now I'm intrigued. Will have to set something up.
Click to expand...
Color/tonality seems to have something to do with it. Using stuffed animals(yes eye detect works on them) the tracking was persistant even with the focus box off of the animal when on a similar color/shade BG. With a different color/shade BG the tracking drops the eye after a second or two. All this with close BG. Tried the same thing with various prints with same result. Similar color BG resulted in persistant tracking on the eye and significantly different color/shade BG resulted in dropped tracking after a brief pause. So apparently as long as the camera "thinks" it's still on the animal it holds the eye but can be confused by similar colors/tones.
 
Member's gallery
NorthernFocus said:
Color/tonality seems to have something to do with it. Using stuffed animals(yes eye detect works on them) the tracking was persistant even with the focus box off of the animal when on a similar color/shade BG. With a different color/shade BG the tracking drops the eye after a second or two. All this with close BG. Tried the same thing with various prints with same result. Similar color BG resulted in persistant tracking on the eye and significantly different color/shade BG resulted in dropped tracking after a brief pause. So apparently as long as the camera "thinks" it's still on the animal it holds the eye but can be confused by similar colors/tones.
Click to expand...

Interesting. Thanks for testing it further. I don't have mine anymore to retest it.
 
You must log in or register to reply here.
Back
Top