[totsmuyco]

Nice shots clackers. Are you using AFS rather than AFC?

[clackers]

Nice shots clackers. Are you using AFS rather than AFC?

Absolutely not, Totsmuyco.

*And* shake reduction is on.

I'm using a monopod in the bikes picture, handholding for the female runner.

[totsmuyco]

The k-1 AF tracking must be good. I still have to have a lot of practice on my k-3II.

[Jeffvan]

That sequence is in fact at f4.5, Jeff, so if I can do it, you can do it.

The focal length is 140mm at the furthest, 78mm as she runs past me.

And if you're shooting as close as I was with as long as that lens, yes, on full frame you're really dicing with Depth of Field.

With a small dog, I think you're going to struggle to keep the AF point over the eyes... it will be very easy for the tail or the collar to be the strongest line under the point at a given time.

Below you can see that f5.6 with the same K-1 and Tammy 70-200 f2.8 isn't deep enough to get all three riders on full frame at 130mm or so - it would need to be more f8, even f11.

The cyclists are approaching my camera at around fifty kilometres per hour (30 mph or so).

Thank you for that information clackers. If you wouldn't mind sharing your autofocus settings, I would greatly appreciate it.

From this image:


I gathered the following information from this picture:
Focal Length: 140mm
F-Stop: 4.5
Runner Height: ~1.5m
Distance from runner: ~20m
With those settings I calculate your DOF to be 5.3m


What I am really curious about is this picture:


Do you have a sequence of these pictures? And do you have the RAW files available? I would love to be able to track the focus through a sequence of shots.

The other possibility that comes back at this stage, is the potential of the lens being a factor. I have the Pentax 70-200 and you are using the Tamron 70-200, which is something I mentioned before as not being able to rule out.
One more question, do you shoot RAW or JPEG? It just popped into my head that this might be some sort of contributor, if it is CPU ralated.

Thank you clackers, I hope you don't mind sharing your setup. It might help me figure out how to improve my setup.

[Jeffvan]

Clackers inspired me to go do more tests. And I have some interesting results.

Again the setup is 70-200 fixed to 200mm and f/2.8. Everything is off, SR, NR, etc. The idea isn’t to get the best picture but to see where the focus lands.

I had an inspiration to test two settings more extensively. The first is to go to Release-Priority on the camera and see if things improve. What I found was that the autofocus is still not very accurate but you now have a new feature that does land you some good shots, once in a while. The jumping focus is now back. Generally the focus starts behind the target and then passes the target. At this point you have a shot at getting a picture right on where you want the focus, if the firing of the shutter is on the spot. The next step is that the autofocus detects the front focus and either stalls the progression or reverses the focus direction. At this stage the focal range passes through your target again and you can yet again have a possibility of a picture you would like. It might cycle through this several times but once in a while it will settle on your target for a few shots, and then it will loose it again.

The picture sequence below demonstrates this (I have several similar sequences):
















In picture 7 it appears as if we have a dead on hit, but the focus is actually mid body of the dog. In picture 10 through 13 we have dead on hits. The focus then falls behind again, from frame 14 to 15. So the in focus shots is 4 out of 15. Which is a lot better than my previous method. There might be something to say about the lens’ ability for fine control at the further distances. The amount of movement on the focus ring from 10 meters to infinity is really small and this could be a culprit. But keep in mind the focus is lost again at frame 14 and 15. So the lens could still be a very viable culprit at this point.

And then I did a second test. This time I switched to JPEG. My thinking was that the JPEG compression algorithms are well established and many dedicated CPUs have built-in pipelines for JPEG compression. Maybe this would free up the CPU and the buffer. I was back to Focus Priority and 200mm with f/2.8.

JPEG with Focus Priority Sequence:
















This time it was 7/ 15 of the pictures that were dead on target, and the other missed ones were really close! And with quite a few of the frames the target was far away, which shows that the Focus Ring throw range hypothesis, from above, is most likely not the culprit. If this is the better method for action shots, then it would be up to the photographer to decide if they want to abandon their 14 bit depth of field for 8 bits and have all their settings baked into their file.

I will try to head out again and get some more shots with the two different methods to see if there is truly something to these methods.
I am not willing to kill my dog for my hobby, so this takes time.

[Jeffvan]

I am posting this as a separate post to keep previous content intact, so that those that read this at a later date may follow the bread crumbs and potentially find possible solutions.

Some new test results:
I selected 4 sets of dog runs with the following settings:
Pentax K1, Pentax 70-200 at 200mm f/2.8. Shutter speed 1/5000 using TAv. File type JPEG. Focus-Priority On. Shake Reduction on. On a tripod. Each Sequence held between 12 to 15 pictures. The average accuracy of the test was 41.15%.

The next test was with another 4 sets of dog runs and a Pentax K1, Pentax 70-200 at 200mm f/2.8. Shutter speed 1/5000 using TAv. File type JPEG. Release-Priority On. Shake Reduction on. On a tripod. Each Sequence held between 13 to 18 pictures. The average accuracy of the test was 23.47%

I determined that the focus was accurate if I found the face of the dog completely in focus. The truth is that some of the shots were still back or front focus in some of the results, even though the face was completely in focus.

I would like to duplicate the Focus-Priority tests again with RAW files, to make sure that I have consistent tests to reference. I would also like to test Full Manual mode as well.
I surmise at this stage: that the best method for action shots is to use JPEG as a file format with Focus-Priority.

[clackers]

Thank you for that information clackers. If you wouldn't mind sharing your autofocus settings, I would greatly appreciate it..

It would be either Spot or Expanded Area from the central point, Jeffvan, AF-C, probably with Hold on Medium for the cyclist, since people leaning over the fence could be obstacles while panning.

That's about it. Beholder has some of his favourite settings at DSLR Guide for Shooting Sports II: Horse Racing - Articles and Tips | PentaxForums.com

There is little evidence that the DPR reviewers knew what they were doing, sadly.

According to them, such shots are impossible.

What I am really curious about is this picture:


Do you have a sequence of these pictures? And do you have the RAW files available? I would love to be able to track the focus through a sequence of shots..

No, but you are welcome to browse another unbroken sequence I did as a test using very old Pentax equipment at:

Pentax K-1: continuous sequences of subjects moving towards the photographer - AF.C - Page 3 - PentaxForums.com

and Biz-engineer did a sequence in the Z-axis of some of the fastest birds in the world at:

Can anyone with a K-1 give me a review of the AF? - Page 2 - PentaxForums.com

One more question, do you shoot RAW or JPEG? It just popped into my head that this might be some sort of contributor, if it is CPU ralated.

Thank you clackers, I hope you don't mind sharing your setup. It might help me figure out how to improve my setup.

I will usually shoot RAW+JPG, back button focus so the act of focusing is separate from shooting. I will start tracking and shooting before the athlete enters the area I really want to capture them in, and also follow them as they go past it.

If I wanted to push the framerate I'd do JPG only.

[clackers]

The k-1 AF tracking must be good. I still have to have a lot of practice on my k-3II.

I think it has the same tracking capability, Totsmuyco ... the real skill is you.

A sports photographer smoothly pans keeping the autofocus point on a high contrast line on the subject.

[tregubovav]

...
keeping the autofocus point on a high contrast line on the subject.

Any kind of phase detection autofocus focus on most contrast object in sensor(s)' zone(s).
Another thing - sensor's zone is larger than marker displayed in viewfinder for this/these selected sensor(s).
In AF-S Auto mode with Release-Priority mode tries to "predict" position of object and periodically correct "prediction" by scanning AF sensors. This why you see more misses than using Focus-Priority mode.

[clackers]

Any kind of phase detection autofocus focus on most contrast object in sensor(s)' zone(s)..

No, that's CDAF.

PDAF looks for a distinct line, usually horizontal, that it can rejoin after splitting its image across the ends of two sensors, like the old rangefinder cameras.

If you're quick thinking, you can use a proxy like a shirt collar instead of the eyes in a backlit situation, for instance.

Another thing - sensor's zone is larger than marker displayed in viewfinder for this/these selected sensor(s). .

Correct, so in the dog pictures above, the AF module never focuses on the dog.

It focuses on a particular line or edge under a particular AF circle on the dog. That can be the tail or the collar, whatever it sees under a point.

I like to remove the camera's guesswork and pick the AF point in advance, like the centre one for an action shot.

[tregubovav]

No, that's CDAF.

Manual still says:
Autofocus System
Type: TTL: phase-matching autofocus

CDAF used only for LiveView mode (PDAF is not work when mirror is locked up).

[clackers]

Manual still says:
Autofocus System
Type: TTL: phase-matching autofocus

CDAF used only for LiveView mode (PDAF is not work when mirror is locked up).

Phase Detection is what we're all shooting here, in these pictures, Tegubovav. It uses the mirror.

Contrast Detection uses the sensor and only operates during Live View, by itself too slow for action.

[Winder]

I am posting this as a separate post to keep previous content intact, so that those that read this at a later date may follow the bread crumbs and potentially find possible solutions.

Some new test results:
I selected 4 sets of dog runs with the following settings:
Pentax K1, Pentax 70-200 at 200mm f/2.8. Shutter speed 1/5000 using TAv. File type JPEG. Focus-Priority On. Shake Reduction on. On a tripod. Each Sequence held between 12 to 15 pictures. The average accuracy of the test was 41.15%.

The next test was with another 4 sets of dog runs and a Pentax K1, Pentax 70-200 at 200mm f/2.8. Shutter speed 1/5000 using TAv. File type JPEG. Release-Priority On. Shake Reduction on. On a tripod. Each Sequence held between 13 to 18 pictures. The average accuracy of the test was 23.47%

I determined that the focus was accurate if I found the face of the dog completely in focus. The truth is that some of the shots were still back or front focus in some of the results, even though the face was completely in focus.

I would like to duplicate the Focus-Priority tests again with RAW files, to make sure that I have consistent tests to reference. I would also like to test Full Manual mode as well.
I surmise at this stage: that the best method for action shots is to use JPEG as a file format with Focus-Priority.

You don't need 1/5000 of a second. 1/1000 should be fast enough to freeze the motion of your dog. The DoF at F/2.8 is going to be your problem. Most of the people showing good results are shooting at F/4 or even F/8. They aren't stressing the AF as much as you are with F/2.8 at 200mm. When the DoF is larger than the subject it a lot easier for the camera to get the head in focus. Change your setting to F/5.6 and give it a try. You will get much better results. The K-1 AF isn't going to give you great results at F/2.8 200mm.

[Jeffvan]

Correct, so in the dog pictures above, the AF module never focuses on the dog.

It focuses on a particular line or edge under a particular AF circle on the dog. That can be the tail or the collar, whatever it sees under a point.

I like to remove the camera's guesswork and pick the AF point in advance, like the centre one for an action shot.

So I measured my Center Spot location and created an Action in PS to place it on the picture. I don't have new pictures, so I am using some of the pictures from before to give examples of where the center spot is.
You will see quite a few contrast lines, which should be enough for the phase detection, within the center spot. I flipped the images to Black and White to show the contrast lines.

Here is one of the first pictures I shared in this thread:


Here is a side by side of the brown dog. There is less contrast lines between his ears and tail than between his nose and eyes. There is a another strong contrast line between the Frisbee and his chest and nose.


But I agree: it is a variable that should be removed from the equation. I will try to setup a more reliable test method to make sure there are no problems related to my shooting style.

You don't need 1/5000 of a second. 1/1000 should be fast enough to freeze the motion of your dog. The DoF at F/2.8 is going to be your problem. Most of the people showing good results are shooting at F/4 or even F/8. They aren't stressing the AF as much as you are with F/2.8 at 200mm. When the DoF is larger than the subject it a lot easier for the camera to get the head in focus. Change your setting to F/5.6 and give it a try. You will get much better results. The K-1 AF isn't going to give you great results at F/2.8 200mm.

Winder thank you for that information. I mentioned above that I am shooting at f/2.8 on purpose. I agree that I can get excellent pictures at f/8. What I am attempting to do is find the best settings for autofocus, for a target moving towards me. This will give me more choices, so that I may select lower F-stop values and trade that for better ISO values and shutter speed values. The 1/5000 choice was just based on the ISO range and brightness at f/2.8 that day. The f/2.8 is a deliberate choice, since it will highlight any flaws in the autofocus system. Furthermore, I would like the option to separate the face of the dog from the background and its body. I noticed a lot of suggestions to increase the DOF.

I am experiencing a lot of focus issues. There are other people that echo my findings. Then on the other hand there are people that experience no autofocus problems. Like clackers, and biz-engineer. In some of their pictures I find large DOFs (which is great if that is what they wanted, but if there are any autofocus flaws in those shots then I can't discern it).
To highlight the autofocus success and failures we need a shallow DOF, compared to the target, so we can find where the autofocus lands. We also need shot sequences to be able to see if there are misses by the autofocus. biz-engineer seems to cover a lot of those bases, but we cannot see where the focus really lands since the whole target is in focus.

So why are some people finding the autofocus lacking and other don't?
I am attempting to isolate the variables, which is difficult because I have only one pentax camera and one 70-200 lens. I am attempting a more scientific approach to the solution, which requires identifying the variables. biz-engineer and clackers might have a better version of the K-1 than what I have, or better lenses than what I have. Those are very possible variables. They might also have a better shooting style or better settings.
To figure this out, I am trying different settings to identify and eliminate variables. I am trying to avoid the woohoo effect where we feel like X and Y is the solution, and instead be more empirical with the results. It is frustrating and tedious on my part, since I am noticing a lot of woohoo, and testing other's settings with no observable change (I do need a better setup to retrieve more empirical results).

A common woohoo I have heard is: Person X didn't know what they were doing and that is why they had bad results with autofocus. I find this doubtful when people who have a lot of camera experience (NOT ME, Other testers) suddenly consistently get bad autofocus results with a pentax. Questions I have are: what settings did they get wrong? what did they need to do differently in their style, that is different from other camera manufacturers? If the solution is a larger DOF, does that really mean the autofocus of the pentax is on par with other camera manufacturers? I want to know as many true things and as few false things as possible. I have no bone to pick here, just a drive for getting the most out of my K-1 and lenses.

I will be working on a better rig to eliminate my shooting style. If I find the best settings for the autofocus system, and I am still experience problems, then the only possibility is that my lens or camera, or both are not on par with what biz-enginner or clackers'. At that stage I might rent another K-1 and a lens and do the same tests and see what I get. Maybe biz-enginner and clackers have exceptional versions of the camera or lenses. Or they actually have similar focusing issues but have been able to overcome the problem with style and DOF tuning. If you go back to the sequence of shots with JPEG and Focus-Priority you will see that the focus is almost exactly on the face, for many of the shots. So yes, a larger DOF would hide this problem completely.

I hope this highlights how many variables are at play, and so far I have found that RAW(DNG) is a factor in the delay, at least in my camera, between autofocus and shutter release. This could be related to the embedded jpeg in the DNG file, or all RAW files have this issue. But I need to do more tests to determine this. But for now I need to take a break (while I travel) and it will take some time to build a rig to test autofocus and eliminate my shooting style as a factor. I am hoping with a rig that runs at a constant speed, I can actually measure the delays and location of the focus more accurately. This will allow for much more precise determination of which factors play an actual role in getting the best autofocus results.

I apologize if anyone is frustrated by my posts. I truly do value any input, which would lead me to get better results and identify more variables.

Thank you!

[Winder]

Winder thank you for that information. I mentioned above that I am shooting at f/2.8 on purpose. I agree that I can get excellent pictures at f/8. What I am attempting to do is find the best settings for autofocus, for a target moving towards me. This will give me more choices, so that I may select lower F-stop values and trade that for better ISO values and shutter speed values. The 1/5000 choice was just based on the ISO range and brightness at f/2.8 that day. The f/2.8 is a deliberate choice, since it will highlight any flaws in the autofocus system. Furthermore, I would like the option to separate the face of the dog from the background and its body. I noticed a lot of suggestions to increase the DOF.

I am experiencing a lot of focus issues. There are other people that echo my findings. Then on the other hand there are people that experience no autofocus problems. Like clackers, and biz-engineer. In some of their pictures I find large DOFs (which is great if that is what they wanted, but if there are any autofocus flaws in those shots then I can't discern it).
To highlight the autofocus success and failures we need a shallow DOF, compared to the target, so we can find where the autofocus lands. We also need shot sequences to be able to see if there are misses by the autofocus. biz-engineer seems to cover a lot of those bases, but we cannot see where the focus really lands since the whole target is in focus.

What I'm saying is that the people who are getting good results are using F/4 - F/8. They shoot where the DoF is as large as the subject. As I have said previously, There is latency in the system which is exactly what you are seeing. The AF locks, but by the time the shutter releases the target point as moved out of the plane of focus when using F/2.8. As you say, using F/4 - F/8 doesn't actually test the AF speed. As you point out Biz-Engineer is shooting his horses are F/4 - F/8 so that most if not all of the horse fits in the DoF. Its not really much of a test for the AF as you could employ zone focusing and get the same result on a manual focus camera.