Struggling with Z8 focus tracking

[SCoombs]

You need to watch the video I linked to. If you are doing significant crops, then you are losing isolation and gaining noise. If it is just small crops you may not notice the loss of isolation or noise, but significant crops you will. Watch the video, it is all explained and demonstrated. A cropped 70-200 to 400mm will be the same DOF as a 400mm f5.6 lens and have the same noise as double your ISO. There is no getting around this.

I would never use my 70-200 f2.8 over my 180-600 for birding (unless it was in a cage or VERY tame!) as it wouldn't be able to pick up the bird's eye for a start, but the resultant crop would be worse. That is the very reason *why* these longer lenses exist, in order to make focusing easier, to make the resultant image IQ better. If you are looking for better IQ at that distance, then you need to invest in a better lens like a 400 f2.8TC, this is the price you pay for wanting great IQ and focal length. Using a 70-200 f2.8 in order to try to achieve this is destined to fail.

You can say there is no getting around it but I have actually taken photos side by side of the same subject and the 70-200 were much, much better because of the difference in noise even with the crop. This wasn't theory only, but it was practice. The difference was the difference. Those two tennis photos are from a day when I had both lenses and I would shoot a player with one, then swap and shoot the other same subject, same location, same light. At the end of the day the 70-200 shots were vastly superior.

And again: not every photo is supposed to be a close up. Whether I'm using a 200mm lens or a 600mm lens, I'm going to want a field of view for many shots which would make the subject "small." Here is one of the most popular shots from a recent game:

You can only see EXIF info for this image if you are logged in.

This is exactly the composition I would want. Any more zoomed in and it's just not the same photo - but these kids are small enough in the frame that the tracking is behaving in the difficult way I've been describing here. This was on my 180-600 at 180mm. I could have zoomed in and gotten a better chance at better focus tracking, but at the expense of not getting this photo. In fact, if I'd had the 70-200 on for this shot it would have been even better because the depth of field would have been much more shallow.

So yes, a longer lens can make a difference, but there is a certain level of zoom here where I am struggling to get the camera to maintain focus but which is exactly as zoomed in as it should be.

I'd also say this: I'm the only photographer (and I'm talking about professionals, not parents/whatever) I ever see at any of these games that even brings a longer lens than the 70-200. I come with the 180-600 in a backpack and occasionally swap to it. The other professionals are always just carrying a camera with a 70-200 on it. Obviously it is different at pro-sports games and college sports games, but this is the norm I see for high school sports, at least around here.

 

[Lance B]

You can say there is no getting around it but I have actually taken photos side by side of the same subject and the 70-200 were much, much better because of the difference in noise even with the crop. This wasn't theory only, but it was practice. The difference was the difference. Those two tennis photos are from a day when I had both lenses and I would shoot a player with one, then swap and shoot the other same subject, same location, same light. At the end of the day the 70-200 shots were vastly superior.

And again: not every photo is supposed to be a close up. Whether I'm using a 200mm lens or a 600mm lens, I'm going to want a field of view for many shots which would make the subject "small." Here is one of the most popular shots from a recent game:

View attachment 99516

This is exactly the composition I would want. Any more zoomed in and it's just not the same photo - but these kids are small enough in the frame that the tracking is behaving in the difficult way I've been describing here. This was on my 180-600 at 180mm. I could have zoomed in and gotten a better chance at better focus tracking, but at the expense of not getting this photo. In fact, if I'd had the 70-200 on for this shot it would have been even better because the depth of field would have been much more shallow.

So yes, a longer lens can make a difference, but there is a certain level of zoom here where I am struggling to get the camera to maintain focus but which is exactly as zoomed in as it should be.

I'd also say this: I'm the only photographer (and I'm talking about professionals, not parents/whatever) I ever see at any of these games that even brings a longer lens than the 70-200. I come with the 180-600 in a backpack and occasionally swap to it. The other professionals are always just carrying a camera with a 70-200 on it. Obviously it is different at pro-sports games and college sports games, but this is the norm I see for high school sports, at least around here.

Sorry, there is no getting round physics and if you watched the video, you would see that Steve has shown irrefutable evidence, and I can attest to it as well as I have also seen it with my own eyes. As Steve points out, at the same subject distance and the same focal length the 70-200 *will* be better due to the f2.8 max aperture. but when you crop, that advantage is lost depending on how much you crop. You are also trying to do all things with one lens which just may not be possible in this case, especially such a short lens like the 70-200 f2.8. Ever wonder why sports photogs are carrying two or three bodies with a different lens attached - for exactly the scenario you are describing that you are trying to do with one lens. In the case above at 180mm, yes you would have a "better" shot with the 70-200 compared to the 180-600 when both at 180mm. However, when cropping those far off shots, that advantage is progressively lost the more you crop. That's fine, but it is at the expense of getting the shot at 200mm when the action is at the other end of the field and missing focus and then cropping to 400mm or longer and wondering why they aren't in focus. The subjects are just too small for reliable AF. When I did soccer matches, I used my 80-400 f4.5-5.6 on my D810/D850/D500 when I shot DSLR's. My photos were excellent quality and the person I did them for was over the moon. Still had isolation and exemplary IQ. I would use my 100-400 these days on my Z8 or Z9 however, I just don't so it anymore.

 

[SCoombs]

Sorry, there is no getting round physics and if you watched the video, you would see that Steve has shown irrefutable evidence, and I can attest to it as well as I have also seen it with my own eyes. As Steve points out, at the same subject distance and the same focal length the 70-200 *will* be better due to the f2.8 max aperture. but when you crop, that advantage is lost depending on how much you crop. You are also trying to do all things with one lens which just may not be possible in this case, especially such a short lens like the 70-200 f2.8. Ever wonder why sports photogs are carrying two or three bodies with a different lens attached - for exactly the scenario you are describing that you are trying to do with one lens. In the case above at 180mm, yes you would have a "better" shot with the 70-200 compared to the 180-600 when both at 180mm. However, when cropping those far off shots, that advantage is progressively lost the more you crop. That's fine, but it is at the expense of getting the shot at 200mm when the action is at the other end of the field and missing focus and then cropping to 400mm or longer and wondering why they aren't in focus. The subjects are just too small for reliable AF. When I did soccer matches, I used my 80-400 f4.5-5.6 on my D810/D850/D500 when I shot DSLR's. My photos were excellent quality and the person I did them for was over the moon. Still had isolation and exemplary IQ. I would use my 100-400 these days on my Z8 or Z9 however, I just don't so it anymore.

One can cite whatever physics are wanted, but I took the photos. Hundreds of them. The proof was in the pudding. Scientific theory is nice and I am a physics teacher so I am not without regard for it, but whereas scientific theories work perfectly in a laboratory or under very specific conditions, reality is a lot more complex than that and there are usually many factors at play causing what is predicted by theory not to pan out in the real world in as pristine a way as they do in controlled conditions. Suffice it to say, the physics of what is theoretically, in ideal conditions supposed to happen is not going to change what actually happened and what I saw with my own two eyes over hundreds of photos worth of examples.

And again, this entire thing is very significantly missing the point. Regardless of what focal length I am using, if I am shooting a photo like the example I posted, I need to be able to have reliable AF. I am not getting reliable AF in those cases, and zooming in with a longer lens doesn't fix that problem: it just hides it by being another version of not being able to get those shots. So sure, I could have turned the zoom ring and taken this photo instead. The focus and sharpness and everything would have been better. It wouldn't have been the photo I wanted, though. In this case, it didn't matter: the AF worked fine.

Yet say the AF had grabbed onto the background in this case as it has in many other cases so I missed the shot. In that case, if I could go back in time and zoom in more so that the AF worked better, would that have changed the fact that I didn't get the shot? No. So the problem remains: reliable AF tracking for shots of this sort.

You can only see EXIF info for this image if you are logged in.

 

[Lance B]

There are obviously other factors at play here regarding correct AF settings not just related to distance and focal length etc as i have pointed out to you. However, it is complicated and many have given you advice that you either want to ignore or find any reason to find fault regardless of suitability of said advice. I give up trying, not because I think I am wrong but because there seems to be a certain reluctance on your part to heed advice. I know you have posted the most complaints regarding AF issues I have ever seen with regards to the Z8 or Z9 and all I can say is it might be time to move on. I bid you farewell.

 

[SCoombs]

There are obviously other factors at play here regarding correct AF settings not just related to distance and focal length etc as i have pointed out to you. However, it is complicated and many have given you advice that you either want to ignore or find any reason to find fault regardless of suitability of said advice. I give up trying, not because I think I am wrong but because there seems to be a certain reluctance on your part to heed advice. I know you have posted the most complaints regarding AF issues I have ever seen with regards to the Z8 or Z9 and all I can say is it might be time to move on. I bid you farewell.

The reality is that if I had to do it over I would probably not choose the Z8. I have long been a Nikon loyalist, but I just haven't been able to get it to work for me. I really very much prefer Nikon when it comes to literally everything besides the AF. I've used Canons, for instance - and some of their lowest, earliest mirrorless options at that - and the AF was simply vastly, vastly superior to what I have personally experienced from the Z8, I am sorry to say. I had a Canon M50 for about a year but I much preferred Nikon for so I sold it and went that direction before starting to really invest in lenses. The lenses I have now are obviously much better - unfortunately Canon never really did much with the EF-M mount - but the AF has been apples and oranges. That M50 just focused exactly where I wanted it to and gave me not the slightest trouble in keeping the focus on whatever it was.

But you're right, I've had significant problems with the AF on the Z8 in spite of putting in enormous amounts of effort in trying to ask advice and in trying to get them sorted out. If I come across as reluctant to heed advice here, it's not because I'm unwilling to take advice or trying to be flippant or rude - and I do apologize for coming across that way if I at all do. Rather, my perceived reluctance is just because the advice is not anything different than what I've gotten and already spent tens of thousands of photos practicing and trying to make work already. For instance, the most common advice here and on the other forum where I have posted this is to use subject detection, but the reality is that I've already spent about 275,000 photos doing that (I just double checked my shutter count) and there are many reasons I started using dynamic area instead.

If money were no factor, I would probably move on at this point. Unfortunately, money is very much a factor and moving on is simply not even a remote possibility, so I have to try to get the best out of what I have. To that end, what I have specifically tried to ask about here is why the dynamic area modes do not appear to be operating the way that they are documented to operate. Specifically, the camera is supposed to keep focus on things which remain under the helper points, at least temporarily, but in actuality as soon as the subject is under a helper point it is immediately refocusing. If it weren't a very specific question like that, I would not have posted because, as you have noted, I've already asked plenty of other more general questions to try to sort this AF out.

 

[Ralph]

The reality is that if I had to do it over I would probably not choose the Z8. I have long been a Nikon loyalist, but I just haven't been able to get it to work for me. I really very much prefer Nikon when it comes to literally everything besides the AF. I've used Canons, for instance - and some of their lowest, earliest mirrorless options at that - and the AF was simply vastly, vastly superior to what I have personally experienced from the Z8, I am sorry to say. I had a Canon M50 for about a year but I much preferred Nikon for so I sold it and went that direction before starting to really invest in lenses. The lenses I have now are obviously much better - unfortunately Canon never really did much with the EF-M mount - but the AF has been apples and oranges. That M50 just focused exactly where I wanted it to and gave me not the slightest trouble in keeping the focus on whatever it was.

But you're right, I've had significant problems with the AF on the Z8 in spite of putting in enormous amounts of effort in trying to ask advice and in trying to get them sorted out. If I come across as reluctant to heed advice here, it's not because I'm unwilling to take advice or trying to be flippant or rude - and I do apologize for coming across that way if I at all do. Rather, my perceived reluctance is just because the advice is not anything different than what I've gotten and already spent tens of thousands of photos practicing and trying to make work already. For instance, the most common advice here and on the other forum where I have posted this is to use subject detection, but the reality is that I've already spent about 100,000 photos doing that and there are many reasons I started using dynamic area instead.

If money were no factor, I would probably move on at this point. Unfortunately, money is very much a factor and moving on is simply not even a remote possibility, so I have to try to get the best out of what I have. To that end, what I have specifically tried to ask about here is why the dynamic area modes do not appear to be operating the way that they are documented to operate. Specifically, the camera is supposed to keep focus on things which remain under the helper points, at least temporarily, but in actuality as soon as the subject is under a helper point it is immediately refocusing. If it weren't a very specific question like that, I would not have posted because, as you have noted, I've already asked plenty of other more general questions to try to sort this AF out.

I‘ve been reluctant to get into this discussion but I am now willing to add my 2 cents for what it is worth. Since your main point is about dynamic mode, my understanding is that the helper points only work with a clean background like a bird in a BIF. I may be wrong but once the main focusing point leaves the subject the camera will only maintain focus if there are no other subjects to grab onto. I don’t see how it could keep focus if the are numerous other subjects in the focusing area.

 

[SCoombs]

I‘ve been reluctant to get into this discussion but I am now willing to add my 2 cents for what it is worth. Since your main point is about dynamic mode, my understanding is that the helper points only work with a clean background like a bird in a BIF. I may be wrong but once the main focusing point leaves the subject the camera will only maintain focus if there are no other subjects to grab onto. I don’t see how it could keep focus if the are numerous other subjects in the focusing area.

Hmm, I hadn't ever seen that. It might explain things a bit, though for what it's worth I went out yesterday to a location with a pretty uniform background to test and still found I could replicate the same behaviors.

To make sure I am clear though, I'm not somehow insistent upon using dynamic area and only dynamic area. I only started doing so after encountering a variety of problems over tens of thousands of shots using other options. In fact for the first 6 months to a year of owning the Z8 and for the entire year and a half before that I owned a Z7ii, I pretty much never used dynamic area modes because the general consensus you'd read was that these were legacy modes and that really you should be using the area modes, which were the way the camera was intended to be used. Then sometime last year I started to see a lot more posts in different places and a lot more videos from "experts" where people had really warmed up to dynamic area modes on Nikon mirrorless and the idea was now that actually they are quite useful for certain things. Meanwhile I'd been encountering a lot of problems with the area modes and so around that time I really stopped using the area modes and subject detection because it just wasn't working very well so reading all of this "newfound wisdom" I tried dynamic area modes and found them to be working a lot better.

All of this is to say that I will use whatever works. The issue I am having is that at this point I've put a lot of time into trying pretty much everything.

 

[fcotterill]

@SCoombs
In your situation, especially considering the vast volume of images taken, plus frustrations, I would compare the Z8 with your preferred lenses - ie focal lengths - against a D6 using C1, C2, also Dynamic modes with 3D mode.
And with Face and Eye detection turned in each AFmode in the D6 autofocus submenu (rent a D6 if necessary)
This comparison should reveal if the stickiness issues, including tracking performance, devolves to the different AF sensor systems.

 

[jboyle9808]

Shane,

I hesitate to respond to these types of discussions because I have very little experience compared to everyone else who is posting, but I had a thought when reading through this thread when you talked about how quickly/erratically the subject was moving. Have you experimented with changing the subject motion in A3 from steady to erratic or vice versa? I apologize if you've already addressed this and I missed it in the thread.

 

[Wes Peterson]

Id get out of dynamic and build a C-1 box the size of a player when they are smallest in frame and turn on human subject detection. at the distance you are shooting and the subject size you can easily give the camera too much info to work with. Also when the camera likes to jump to the background the blocked shot response should be set to quick. You want the AF to jump to the subject ASAP when you get the subject under the af point

 

[ajrmd]

Id get out of dynamic and build a C-1 box the size of a player when they are smallest in frame and turn on human subject detection. at the distance you are shooting and the subject size you can easily give the camera too much info to work with. Also when the camera likes to jump to the background the blocked shot response should be set to quick. You want the AF to jump to the subject ASAP when you get the subject under the af point

I do this frequently on site depending on the subject, setting, etc. Sometimes it seems to work, at other times it is less successful which points back to the fundamental questions...

 

[SCoombs]

Shane,

I hesitate to respond to these types of discussions because I have very little experience compared to everyone else who is posting, but I had a thought when reading through this thread when you talked about how quickly/erratically the subject was moving. Have you experimented with changing the subject motion in A3 from steady to erratic or vice versa? I apologize if you've already addressed this and I missed it in the thread.

Yes. That's actually the one setting I can't detect any difference with. At least with the 1- 5 setting I can see a clear difference in behavior when I make sure I get a very large subject in frame. I am not saying that it is not doing anything, but whatever it is doing I haven't found a way to "force" the behavior to manifest through any kind of test, and I don't notice a difference in actual use.

 

[SCoombs]

Id get out of dynamic and build a C-1 box the size of a player when they are smallest in frame and turn on human subject detection. at the distance you are shooting and the subject size you can easily give the camera too much info to work with. Also when the camera likes to jump to the background the blocked shot response should be set to quick. You want the AF to jump to the subject ASAP when you get the subject under the af point

Yes, I have done this a lot. As I said, I didn't start in dynamic: I started using it only after seeing the shortcomings over thousands and thousands of photos worth of experience using the area modes, including custom ones.

One big problem I have found with them is that when players start to overlap in the frame, the area modes become a liability because of how closest subject priority and subject detection work. For instance, I have a photo where a player's entire body is clear and fairly large in the frame and I was shooting with a custom area mode on him - BUT the arm of the player defending him was in stretched out in front of his chest, and so the camera focused on the defender instead of the player that the box was on. I have a nice sharp view of the back of someone's head but a somewhat soft face on the player I was trying to photograph!

The other is that whereas the purpose of dynamic area modes is to provide a margin of error when trying to track fast, unpredictably moving subjects, the custom area modes don't. Now in fairness the point of this discussion has largely been the dynamic area modes NOT actually operating with that margin of error as the are theoretically supposed to, but regardless the custom area modes don't even theoretically provide a margin of error and require perfect precision, which is not always achievable even for the best of us.

 

[SCoombs]

@SCoombs
In your situation, especially considering the vast volume of images taken, plus frustrations, I would compare the Z8 with your preferred lenses - ie focal lengths - against a D6 using C1, C2, also Dynamic modes with 3D mode.
And with Face and Eye detection turned in each AFmode in the D6 autofocus submenu (rent a D6 if necessary)
This comparison should reveal if the stickiness issues, including tracking performance, devolves to the different AF sensor systems.

For what it is worth, I shot with a D500 before moving to the Z system and I did not experience a stickiness issue.

 

[Wes Peterson]

Yes. That's actually the one setting I can't detect any difference with. At least with the 1- 5 setting I can see a clear difference in behavior when I make sure I get a very large subject in frame. I am not saying that it is not doing anything, but whatever it is doing I haven't found a way to "force" the behavior to manifest through any kind of test, and I don't notice a difference in actual use.

The steady or erratic setting determines how the AF tracking behaves when the original subject is blocked by something temporarily. If the setting is steady then the lens will continue to track the direction and speed until the AF can see the subject clearly. If set to erratic it will give up the track as to not overshoot. This setting pertains to changes in focal distance only, not "tracking" across the frame.

 

[Wes Peterson]

One big problem I have found with them is that when players start to overlap in the frame, the area modes become a liability because of how closest subject priority and subject detection work. For instance, I have a photo where a player's entire body is clear and fairly large in the frame and I was shooting with a custom area mode on him - BUT the arm of the player defending him was in stretched out in front of his chest, and so the camera focused on the defender instead of the player that the box was on. I have a nice sharp view of the back of someone's head but a somewhat soft face on the player I was trying to photograph!

This is where the delayed blocked shot response setting is supposed to help.

 

[SCoombs]

This is where the delayed blocked shot response setting is supposed to help.

"Supposed" is, I suppose, the key word for this entire discussion. I know what it is supposed to do, and I posted my question because it is not doing what it is supposed to do.

Now to repeat, I actually can get the blocked shot response setting to make a clear difference if the subject is large enough for a headshot. Usually in sports, wildlife, and other forms of photography we want something wider so as to show the entire subject along with some context of the setting. It's when shooting at these more normal sizes that the setting doesn't seem to be doing a lot.

And with the subject detection modes, I am finding that it makes even less of a difference than that.

This having been said, even if it works perfectly it's only a blocked shot if the arm comes in front of the other player after acquiring focus. When the arm is already there from the outset, it means that the custom area method will pretty much never get focus on the desired subject. That's one of the problems I have with it. So for instance, one of the main things that got me to stop using this method was trying to get focus on a goalkeeper with a lot of players converging on the net. In a shot like this one, even if you can see the goalkeeper it is very easy for other people to cover up part of him, and so if you have one of those narrow custom area boxes people use to follow players around the field, it can be a big problem.

You can only see EXIF info for this image if you are logged in.

 

[dabhand16]

One big problem I have found with them is that when players start to overlap in the frame, the area modes become a liability because of how closest subject priority and subject detection work.

This is one of the issues with using a shorter lens. That is the exact reason that I never use any tracking or recognition modes. My now preferred AF is a large single spot. with no other tools. I use BBF and when have focus I'll often release the button to lock the focus and only hit it again when I need to. Sure, I get failures but my keeper rate is way more than adequate.

With team sports it is rare to get isolated players even with longer lenses unless you shoot where there is no action. For me the 100-400 (or 450 equivalent on my Fuji 70-300) is a sweet spot.

 

[SCoombs]

This is one of the issues with using a shorter lens. That is the exact reason that I never use any tracking or recognition modes. My now preferred AF is a large single spot. with no other tools. I use BBF and when have focus I'll often release the button to lock the focus and only hit it again when I need to. Sure, I get failures but my keeper rate is way more than adequate.

With team sports it is rare to get isolated players even with longer lenses unless you shoot where there is no action. For me the 100-400 (or 450 equivalent on my Fuji 70-300) is a sweet spot.

I'm afraid I don't follow why you say that a longer lens means that players getting in front of one another does not obscure the AF in the same way.

 

[dabhand16]

I'm afraid I don't follow why you say that a longer lens means that players getting in front of one another does not obscure the AF in the same way.

They sometime do! But a lot of the time they don't.

I think that you will be spending a lot of your life chasing a theoretical dream. For instance, Steve has said that sometimes bird detection does not work too well so has found that in those circumstances animal detection works better. He has not spent time trying to figure out why the bird detection is not doing what it says on the tin, he has found something that does the job.

 

[SCoombs]

They sometime do! But a lot of the time they don't.

I think that you will be spending a lot of your life chasing a theoretical dream. For instance, Steve has said that sometimes bird detection does not work too well so has found that in those circumstances animal detection works better. He has not spent time trying to figure out why the bird detection is not doing what it says on the tin, he has found something that does the job.

Well I agree with this mindset. I've tried to make it clear that I don't care which mode I use as long as it works. I'm where I'm at with this question because I've shot almost 300,000 photos over all the different modes and found the custom areas not to work as well as dynamic area, but even the dynamic area is not actually keeping focus on subjects the way it is supposed to work. So in other words, I'm here because of just trying to use what works regardless of why it works, but I'm at the end of the line in terms of having other things to try that may work better.

 

[dabhand16]

Well I agree with this mindset. I've tried to make it clear that I don't care which mode I use as long as it works. I'm where I'm at with this question because I've shot almost 300,000 photos over all the different modes and found the custom areas not to work as well as dynamic area, but even the dynamic area is not actually keeping focus on subjects the way it is supposed to work. So in other words, I'm here because of just trying to use what works regardless of why it works, but I'm at the end of the line in terms of having other things to try that may work better.

A bit left field, but how about hiring or borrowing a 100-400? TBH much as I love my 70-200, it is not the tool for field sports. I'm not saying that it will be the silver bullet, but it will give you another perspective and you can compare the hit rates and it could reduce the post processing with less cropping.

On average I shoot about 350-450 images in a rugby match using 7 or 8 fps most of the time and 15 fps for the spot kicks. I then deliver between 120 and 170 image to the club the next day. Many that I cull are good technically but I only keep the best ones.

 

[fcotterill]

"Supposed" is, I suppose, the key word for this entire discussion. I know what it is supposed to do, and I posted my question because it is not doing what it is supposed to do.

Now to repeat, I actually can get the blocked shot response setting to make a clear difference if the subject is large enough for a headshot. Usually in sports, wildlife, and other forms of photography we want something wider so as to show the entire subject along with some context of the setting. It's when shooting at these more normal sizes that the setting doesn't seem to be doing a lot.

And with the subject detection modes, I am finding that it makes even less of a difference than that.

This having been said, even if it works perfectly it's only a blocked shot if the arm comes in front of the other player after acquiring focus. When the arm is already there from the outset, it means that the custom area method will pretty much never get focus on the desired subject. That's one of the problems I have with it. So for instance, one of the main things that got me to stop using this method was trying to get focus on a goalkeeper with a lot of players converging on the net. In a shot like this one, even if you can see the goalkeeper it is very easy for other people to cover up part of him, and so if you have one of those narrow custom area boxes people use to follow players around the field, it can be a big problem.

View attachment 99521

The athletes are too similar, too close together and moving too fast for an automated tracking system.

This scene is extremely challenging for the Autofocus system of any modern camera to work reliably (but it will be interesting to test a D6, which has a very different AF system from the D5 in my experience with both cameras).

Although I don't photograph sports, the only solution is to try and snapshot the face of one of the players using a tight Custom Area in high speed bursts..... trying to keep the head in the AF frame as action kicks in in front of the posts.

Using a custom area mode, hit rate will rely mostly on how much you keep the AF search area on the subject. Hit and miss certainly , with hit rate dependent on the photographer's skill helped by high frame rate. This tactic appears to be what's been suggested by this photographers in your parallel dpr thread.

 

[fulltimewanderer]

Coombs, with all that you have tried and all that has been suggested have you sent the camera and lens or considered sending the camera and lens to Nikon ? After all you have been able do get satisfactory results with other gear, just my two cents worth.

 

[SCoombs]

A bit left field, but how about hiring or borrowing a 100-400? TBH much as I love my 70-200, it is not the tool for field sports. I'm not saying that it will be the silver bullet, but it will give you another perspective and you can compare the hit rates and it could reduce the post processing with less cropping.

On average I shoot about 350-450 images in a rugby match using 7 or 8 fps most of the time and 15 fps for the spot kicks. I then deliver between 120 and 170 image to the club the next day. Many that I cull are good technically but I only keep the best ones.

I have considered a 100-400 and am still considering it but the reality is:

1) I have a 180-600 which I have used and, in spite of insistence of Lance (a guy who deserves a lot of respect, I would add) above, the when the rubber has hit the road it just hasn't remotely compared to the 70-200 even with heavy cropping. Part of this is probably because of the overall sharpness difference between the two lenses, and on that note it's actually maybe a little unclear whether a 100-400 would be an improvement because if I'm not mistaken Steve and others have found the 180-600 to actually be sharper at 300-400mm. The bigger thing, though has been the noise levels. Sure, in theory cropping 200mm at f2.8 should be equal noise-wise to 400mm at f4, but for one thing the 400mm is not at f4 - it's at f5.6 or f6 - and for another, the bottom, rubber hits the road results have been that cropping the 200mm f2.8 has just been better and it hasn't been close, at least when not in great light.

And that's the thing - shooting these HS sports, especially in the autumn, I am often in light that is pushing ISOs up to 6400 or 10,000 shooting at the necessary shutter speed at f5.6 or f6.3, and while modern AI denoising is amazing there is still a point where there isn't enough detail to recover so it doesn't ultimately work as well as just getting more light on the sensor in the first place. The 100-400 would put me at f5.6 on the long end, so I am very skeptical that if the 180-600 looks worse than the 70-200 in the conditions I have to shoot in that the 100-400 would be any better.

2) A big part of the issue here is really coming down to subjects which are already the size I want them in the frame. Yes, more range can make a difference for shots which I am needing to crop a lot, but if the issues I am having are showing up when subjects are already the right size, a longer zoom won't change that: it will just give me more opportunities with subjects at that proper size in the frame where the AF can still have the same problems.

3) I'd love to invest in better equipment, but there are financial considerations here and I think a lot of it can come down to the "tier" of work being done. I forget if I mentioned it earlier in this thread or if it was on another forum, but I'm the only one I ever see at these games that has even brought a lens longer than 70-200. I bring my 180-600 even if I use it much less frequently than the 70-200. The other professionals at these high school games just come with a 70-200. It's very much the norm. I'm not saying that this means I shouldn't use better equipment just because others are not: I'm rather simply saying that the equipment I'm using seems to be the norm, at least around here, for people doing this type of work (vs. professional or college level stuff) and dealing with the financial considerations involved.