[MightyMike]

So I went back to the dog park, there were dozens more dogs as it had rained the whole weekend before so I got fewer runs in. However this time I set AF Hold to Medium and the AF was set to Auto-9-Point, it was sunny and I increased the shutter speed to 1/4000th sec, however I forgot to turn SR off which could lead to unintentional blur if one is panning even with a high shutter speed, I've seen it in airshow photos, for a while it leaves you scratching your head and then you realize SR could have played a role (Certainly not always).

Sorry I didn't produce the photo streams this time however the format is the same as the first test seen here The Ugly Truth About Pentax AF-C (Cuteness included) - PentaxForums.com

1st sequence APS-C mode 18 photos in 4-5 seconds running straight towards me, into shade then out of shade then getting way too close
Ratings (4,2,1,2,4,1,1,1,2,2,1,2,4,2,2,4,4,1)
33.3% Perfect, 72.2% the camera did its job (Combined #1 and #2)
Obviously a weak start to this sequence. The camera managed to track reasonably well when entering the shade, not perfectly but still followed in general, however when exiting the shade the camera hesitated and when it re-locked on it was to the dogs tail and then lost it again until the last shot when the dog was almost on top of me and it managed a shot in focus but by then the dog wasn't able to fit in the frame.

2nd sequence APS-C mode 22 photos in 4-5 seconds not running directly towards me but on an angle and as he gets closer it goes to side on then too close where I'm not able to get him entirely in the frame.
Ratings (1,1,1,1,4,1,1,1,1,4,1,4,4,1,2,4,1,1,1,1,1,1)
72.7% Perfect, 77.7% the camera did its job (Combined #1 and #2)
Obviously the camera tracks less direct Z-axis movement a lot better, however as the dog got closer the cameras tracking started becoming inconsistent but once it locked on confidently again it stayed with the dog despite not being able to keep the dog fully inside the frame.

3rd sequence APS-C mode 27 photos in 5-6 seconds, straight on started in shade for the first shot with a quick transition to sun and back into shade for the last 5 shots but the dog got super close again by that point.
Ratings (1,2,4,4,1,1,2,4,1,2,1,1,1,1,4,2,4,1,1,4,1,2,2,4,4,4,4)
40.7% perfect, 62.9% the camera did its job (Combined #1 and #2)
It seems through several of these the lens would jump ahead then reacquire, track for a bit, lose it and jump ahead then reacquire again. If we were to take out the ending of the sequence to this where I could have stopped shooting sooner it would be more like 52.4% Perfect and 71.4% the camera did its job.

4th sequence APS-C mode 23 photos in 4-5 seconds coming straight at me with an obstruction (another dog) entering the frame part way, despite being on focus hold medium the camera did lose track while that happened.
Ratings (4,2,1,1,1,4,4,1,2,4,4,1,1,1,1,1,2,4,4,4,4,4,2)
39.1% perfect, 56.5% the camera did its job (Combined #1 and #2)
I'd like to point out as in the previous sequence if we take out the bad end we get much better figures 56.3% Perfect and 68.8% the camera did its job.

In total there are 90 photos in 4 sequences, 46.7% perfect and 66.7% the camera did its job. if we leave out those 2 bad endings where I really should have stopped shooting we get 77 photos in 4 sequences, 54.5% perfect and 72.7% the camera did its job. I really can't say Auto-9-Point did any better than SEL33 (Expanded Area 33)

It was interesting to note however that subject transition between sunlight and shade could cause the AF to loose the subject. Maybe this is related to focus hold, maybe its simply a metric the camera has trouble compensating for.

I'm sure I'll post another gallery in the coming days for your enjoyment.

[77me]

Curious if you ever compare what you think is in focus to what the camera thinks is in focus?

[MightyMike]

Curious if you ever compare what you think is in focus to what the camera thinks is in focus?

All the time!!! I try hard to give the camera any excuse possible and many times its clear the camera just chose something other than what I wanted, most of the time this is clear and I really can't blame the camera, its not attached to a super computer server farm with hyper fast and accurate image recognition nor can it read minds. Those who don't realize this are the ones who most regularly complain that they couldn't get any good results out of the system and eventually jump ship. Now I complain that it could be better but I don't complain that it doesn't work because it does work, I have hundreds of thousands of photos with Pentax cameras that prove that (not exaggerating).

Simply put always look for what else the camera may have chose to focus on, you may find out its doing a great job, just not the job you wanted it to do.

[77me]

All the time!!! I try hard to give the camera any excuse possible and many times its clear the camera just chose something other than what I wanted, most of the time this is clear and I really can't blame the camera, its not attached to a super computer server farm with hyper fast and accurate image recognition nor can it read minds. Those who don't realize this are the ones who most regularly complain that they couldn't get any good results out of the system and eventually jump ship. Now I complain that it could be better but I don't complain that it doesn't work because it does work, I have hundreds of thousands of photos with Pentax cameras that prove that (not exaggerating).

Simply put always look for what else the camera may have chose to focus on, you may find out its doing a great job, just not the job you wanted it to do.

I also have many thousands of pictures using the sigma 100-300 f4 both with and without the 1.4 TC. I only occasionally check what the camera thinks is in focus only because I have the first frame set to focus priority and sometimes there is nothing in focus. I mostly take pictures of kids playing sports, so not as fast as dogs. My only complaint is when the camera has a lock, but then for some unknown reason decides to focus on something far away. Sometimes it is my fault for not keeping the subject centered, but not always. On the other hand sometimes I get great unplanned shots of the faces of the parents. I don't doubt that other cameras have better focus tracking, but there is no comparison from k10d to k1, and I have gotten better too.

[MightyMike]

I also have many thousands of pictures using the sigma 100-300 f4 both with and without the 1.4 TC. I only occasionally check what the camera thinks is in focus only because I have the first frame set to focus priority and sometimes there is nothing in focus. I mostly take pictures of kids playing sports, so not as fast as dogs. My only complaint is when the camera has a lock, but then for some unknown reason decides to focus on something far away. Sometimes it is my fault for not keeping the subject centered, but not always. On the other hand sometimes I get great unplanned shots of the faces of the parents. I don't doubt that other cameras have better focus tracking, but there is no comparison from k10d to k1, and I have gotten better too.

You and me both with the Sigma 100-300mm F4.0 EX DG with and without and 1.4x TC. Before focus hold it was a pain in the but, I have one shot of the Snowbirds performing and a seagull far smaller than the group of planes enters the frame and boom the focus jumps before finding its way back. Focus hold is certainly a helpful tool. I started with a K10D and back then it was impressive, I picked one up recently to do SDM changes if and when necessary and OMG its horrific LOL amazing how things have changed.

[77me]

You and me both with the Sigma 100-300mm F4.0 EX DG with and without and 1.4x TC. Before focus hold it was a pain in the but, I have one shot of the Snowbirds performing and a seagull far smaller than the group of planes enters the frame and boom the focus jumps before finding its way back. Focus hold is certainly a helpful tool. I started with a K10D and back then it was impressive, I picked one up recently to do SDM changes if and when necessary and OMG its horrific LOL amazing how things have changed.

Interesting. I have been having more success with focus hold off while at the same time doing a better job of keeping the target centered. Lately I have been switching focus hold between off and low. On the k3 I was using mostly medium.

[MightyMike]

Interesting. I have been having more success with focus hold off while at the same time doing a better job of keeping the target centered. Lately I have been switching focus hold between off and low. On the k3 I was using mostly medium.

Most seem to suggest with focus hold set to off or low one can get better results for an object coming towards you, however when other objects can get in the way focus hold is very helpful. Maybe next time I'm at the dog park I'll shoot with no focus hold and see how it goes. I may even switch to center point only which I suspect will yield even better results so long as I can hold the target in the center.

[Bruce Clark]

All the time!!! I try hard to give the camera any excuse possible and many times its clear the camera just chose something other than what I wanted, most of the time this is clear and I really can't blame the camera, its not attached to a super computer server farm with hyper fast and accurate image recognition nor can it read minds. Those who don't realize this are the ones who most regularly complain that they couldn't get any good results out of the system and eventually jump ship. Now I complain that it could be better but I don't complain that it doesn't work because it does work, I have hundreds of thousands of photos with Pentax cameras that prove that (not exaggerating).

Simply put always look for what else the camera may have chose to focus on, you may find out its doing a great job, just not the job you wanted it to do.

One of the truest statements I have read regarding auto focus. Well done

[MightyMike]

One of the truest statements I have read regarding auto focus. Well done

Thanks Bruce