[MightyMike]

We all know Pentax AF-C is not perfect, we all know its not up to par when compared with the other systems, that really doesn't need to be said over and over again. However just how bad or how good is it? Well the correct answer is YMMV, A skilled photographer will likely get more keepers than a none skilled photographer. The more patience, effort and concentration into achieving the best out come the more keepers you'll get. The more you avoid situations the camera will fowl up in the more keepers you'll get. The better the lens on the camera the more keepers you'll get. But lets face it, we're not all perfect nor do we all own the best gear. I've said it before as a joke with some truth in it, the more bad photos I get the longer I can stay out and shoot for and the less I have to edit when I get home. Sure I do want lots of good photos, but I also want to enjoy just being out and taking pictures too.

So the following is 12 examples of focus tracking on my dog at the dog park on a light overcast afternoon. Now there are some things you may want to know. I shot with the K-1II, TAv, 1/2000th sec (probably should have been faster), F8.0 once but F4.0 the rest of the time, in APS-C and FF modes. AF-C, AF Hold set to low, SEL25 (could have used Auto9). However another thing that may be important to know and you'll see it in the images is my dog is like a shaggy pile of fur with few features to lock onto and track, therefore if the camera got any part of that fur ball running it should be considered a win if not a perfect outcome. I figured when the dog was running full speed it was going about 25-30km/h (15.5-18.8mph) and of course at close range. I also zoomed while shooting therefore the dog will not appear to be getting larger in the frame as he approaches until such a point where I can't zoom out any further. I assure you he covered a lot of ground in a very short amount of time.

I'll use this as my rating system

1) Perfect, spot on focus on the face (or sideways on the body)
2) The camera got some part of the dog in focus (This could be subjective as to whether the camera+lens didn't keep up or whether it just chose a different part of the fur ball to track)
3) The camera failed but that could be because the dog got too close, or I couldn't follow the dog and the dog left the AF area (this could be subjective)
4) Total fail, the camera clearly lost focus for no good reason

Each sequence will be in 2 photo streams, the first will be the series of the full photos resized to 900x600 pixels each for APS-C or 1200x800 pixels each for FF and sharpened. The second stream will be of 800x800 pixel crops for APS-C and 1200x800 pixel crops for FF, no extra editing. The crop sequences don't always show whats in focus for obvious reasons. With each stream I'll include a series of numbers (1-4) as shown above rating each frame.

1st sequence APS-C mode 15 photos in 4-5 seconds, the camera locked onto another dog during the sequence and then back so it still got perfects even if it wasn't the intended dog
Pentax F 70-210mm F4-5.6 @ F8.0 (extra deep hood)
Ratings (4,4,2,4,1,2,2,2,4,2,1,1,1,1,1)
40% perfect ratio (however more so as the dogs stopped moving much on the Z-axis, 73.3% the camera did its job ratio (combined #1 and #2)
Full Frame [8.8mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P01916-01930full.jpg
Crops [6.3mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P01916-01930crop.jpg

I switch now to my preferred lens the Sigma 100-300mm F4.0 EX DG @ F4.0 used this way for the rest of the sequences

2nd sequence APS-C mode 18 photos in 3-4 seconds full head on Z-axis run
Ratings (1,2,4,1,4,2,2,1,1,1,2,4,4,2,2,3,2,2)
Just 27.8% perfect, 72.2% the camera did its job (combined #1 and #2)
Full Frame [7.5mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P01944-01961full.jpg
Crops [5.4mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P01944-01961crop.jpg

3rd sequence APS-C mode 20 photos in 4-5 seconds full head on Z-axis run
Ratings (4,2,4,2,1,1,1,1,1,4,1,2,4,4,4,4,4,3,3,3)
Just 30% perfect, and only 45% the camera did its job (combined #1 and #2) what appears to have happened here though was the camera just couldn't keep up, it was following but from behind the dog and in once instance it got ahead of itself.
Full Frame [8.4mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P01962-01981full.jpg
Crops [6.0mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P01962-01981crop.jpg

4th sequence FF mode 20 photos in 7-8 seconds, starts as head on transitions to side run then to running away from me.
Ratings (2,2,2,1,2,4,1,4,1,2,1,2,1,4,4,1,1,1,2,1)
45% perfect, 80% the camera did its job (combined #1 and #2)
Full Frame [12.9mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P01982-02001full.jpg
Crops [9.1mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P01982-02001crop.jpg

5th sequence FF mode just a short 8 photo burst in 2 seconds head on
Ratings (4,2,4,4,2,1,1,1)
37.5% perfect, 62.5% the camera did its job, rough start to this one but good finish
Full Frame [5.4mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02002-02009full.jpg
Crops [3.6mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02002-02009crop.jpg

6th sequence APS-C mode 12 photos in 2-3 seconds head on
Ratings (4,2,1,1,1,1,4,2,3,3,3,3)
33.3% Perfect, 50% the camera did its job, 33.3% too close!
Full Frame [5.3mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02010-02021full.jpg
Crops [3.8mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02010-02021crop.jpg

7th sequence APS-C mode 14 photos in 3-4 seconds head on, focus hold could have been set to higher. the first few photos didn't capture the dog due to interference from easier to focus on items behind the dog so I gave them a rating of 2 as the camera did its job it just didn't know what it was supposed to do.
Ratings (2,2,2,1,1,1,2,1,1,4,1,1,1,3)
57.1% Perfect, 85.7% the camera did its job
Full Frame [8.3mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02022-02035full.jpg
Crops [5.0mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02022-02035crop.jpg

8th sequence APS-C mode 12 photos in 2-3 seconds towards me to sideways to away from me
Ratings (2,1,4,1,2,2,1,1,4,4,4,1)
41.7% Perfect, 66.7% the camera did its job
Full Frame [4.9mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02054-02065full.jpg
Crops [3.6mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02054-02065crop.jpg

9th sequence APS-C mode 15 photos in 2-3 seconds, head on to sideways to too close. I gave a couple of #1 ratings when the camera maintained focus on the other dog towards the end.
Ratings (2,4,2,4,2,2,1,2,2,1,2,2,1,1,3)
just 26.7% perfect, however 80% the camera did its job
Full Frame [6.3mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02066-02080full.jpg
Crops [4.6mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02066-02080crop.jpg

10th sequence APS-C mode 14 photos in 3-4 seconds slower trot speed relatively head on
Ratings (4,2,1,1,2,1,1,1,1,1,1,1,1,1)
78.6% Perfect, 92.9% the camera did its job, clearly the slower pace of a trot is something the camera can handle pretty well!
Full Frame [6.4mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02086-02099full.jpg
Crops [4.9mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02086-02099crop.jpg

11th sequence APS-C mode 14 photos in 3-4 seconds full speed head on with a turn
Ratings (1,2,4,1,2,1,1,4,2,1,4,1,1,4)
50% perfect, 71.4% the camera did its job
Full Frame [5.6mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02127-02140full.jpg
Crops [4.8mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02127-02140crop.jpg

12th and final sequence APS-C mode just 8 photos in less than 2 seconds head on, I'll award #1 ratings even if the camera transitioned to the other dog.
Ratings (1,2,1,1,1,1,4,1)
75% Perfect (even if it was tracking the wrong dog) 87.5% the camera did its job.
Full Frame [3.2mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02157-02164full.jpg
Crops [2.9mb] http://www.michaelfastphotography.com/galleries/k-1ii/afdog/_K2P02157-02164crop.jpg

Over a spread of 12 sequences, 168 photos the camera got 44% perfect and 72.6% were the camera did as I believe it was supposed to. The camera is certainly far from perfect in this task but it can still provide dozens of good photos over a few short bursts. What is one to do with 74 effectively perfect action photos of their dog (ok and of other dogs too). In my opinion despite all the failures the keepers are well worth it.

Something I've known for a long time and maybe this will help others is that if the first photo in the burst is out of focus the camera will have a harder time re-acquiring focus and will definitely lead to fewer in focus photos in the sequence and a lower keeper rate. However I totally understand that unless you've got amazing eyes and a subject that is easy to see as in or out of focus it will be very hard to know if you're in focus when you start to burst. Another thing clearly demonstrated in these examples is as the subject gets too close the camera (most cameras) can't keep up anymore, if you're smart enough to stop shooting before they get too close your keeper percentage will be higher. The camera also does seem to fail when the Z-axis pace changes or the subject quickly changes direction, do you then fault it? after all it has no idea what the subject is or what it will do next.

So final thoughts... Take this evidence and make of it what you will, not every subject will produce the same results and its clear their is a wide range of results from just one subject. All in all YMMV and the K-1II is not as good as the competition is the most accurate description of its AF-C abilities. I suppose if its not good enough and you aren't willing to wait for better then just jump ship, if you're stubborn like myself keep enjoying shooting until better comes and then enjoy it just as much or more.

---

Last edited by BigMackCam; 06-24-2018 at 12:47 PM.

[jbinpg]

Hi, Mike. See if you can get your hands on the DA 55-300 PLM and redo your test with it. Guaranteed the keeper ratio will be higher.

[MightyMike]

Hi, Mike. See if you can get your hands on the DA 55-300 PLM and redo your test with it. Guaranteed the keeper ratio will be higher.

I know its supposed to be much faster, however don't discount the speed of the Sigma 100-300mm F4.0 EX DG, the only thing going against it is its not HSM so it doesn't keep up with subtle movements as well... that said there is also the DOF advantage of the 55-300 PLM... maybe if one comes my way cheap I'll add it to my collection (that is if I like it). As for redoing the test with a different lens, maybe you want to consider that photographing, editing, evaluating and writing the report took 9-10 hours... I'll have to be in such a mood to do it again.

[normhead]

That's nice but without comparison's from Nikon , Canon and Sony cameras, we have no idea what those numbers mean. I've heard Canon AF called fast but inaccurate, so, I wouldn't in any way concede that Pentax AF is poorer than anyone else's without direct comparisons.

Of course the review sites who are the only ones that could easily do that kind of test, don't. The Camera store Nikon shooter in one of their videos was impressed with Pentax AF. He didn't say it was better, but it was in the ball park. That after a day of shooting all three systems. We don't hear that often. So it's always surprising when a guy doing comparison tests says, "Pentax isn't so bad." Funny how we don't have any Pentax shooters saying that. I think in terms of AF, there's been a lot of hoodwinking going on.

---

Last edited by normhead; 06-23-2018 at 06:16 PM.

[stevebrot]

My level of satisfaction is aided by the fact that I don't own a dog...


Steve

[MightyMike]

That's nice but without comparison's from Nikon , Canon and Sony cameras, we have no idea what those numbers mean. I've heard Canon AF called fast but inaccurate, so, I wouldn't in any way concede that Pentax AF is poorer than anyone else's without direct comparisons.

Of course the review sites who are the only ones that could easily do that kind of test, don't. The Camera store Nikon shooter in one of their videos was impressed with Pentax AF. He didn't say it was better, but it was in the ball park. That after day shooting all three systems. We don't hear that often. So it's always surprising when a guy doing comparison tests says, "This isn't so bad." Funny how we don't have an Pentax shooters saying the. I think in terms of AF, there's been a lot of hoodwinking going on.

I'm basically saying the results aren't that bad, I'm happy with them but always want better. I'm just making it clear what one could expect in such a "real world" situation.

[stillshot2]

I used to shoot the Sony alpha system with the a55 and a330 and various lenses a few years ago and switched to Pentax for reasons different than autofocus. I noticed that I was getting a slightly higher rate of photos in focus with my Pentax K30, regardless of the a55's autofocus system being advanced for its time. My K200D I have now is about on par with the a330 I had. I've also rented high end Nikon gear and have borrowed relatives Canon gear. I use proper technique, and shoot a variety of subjects. There is nothing wrong with Pentax autofocus, you get misses and keepers with any system. It is becoming pathetic how many people are whining about autofocus deficiencies.

[MightyMike]

I used to shoot the Sony alpha system with the a55 and a330 and various lenses a few years ago and switched to Pentax for reasons different than autofocus. I found right away that I was getting a higher rate of photos in focus with my Pentax K30, regardless of the a55's autofocus system being advanced for its time. My K200D I have now is about on par with the a330 I had. I've also rented high end Nikon gear and have borrowed relatives Canon gear. I use proper technique, and shoot a variety of subjects. There is nothing wrong with Pentax autofocus, you get misses and keepers with any system. It is becoming pathetic how many people are whining about autofocus deficiencies.

I totally agree, I hope you don't take my report as whining, Yes I always want better but I'm simply showing what is possible or if you take it the other way what is not possible with Pentax. No hiding the bad results and showing there are still plenty of good results regardless of the bad results.

[photoptimist]

Interesting test, but......

The AF system also has an operating shutter speed that is linked to the aperture of the lens and brightness of the scene (it's in the EXIF data if you care to look). It's quite possible that AF array readings on the rapidly rippling fur, flying dogs, rapid panning, and rapid zooming were simply too blurry for the AF system to read sufficient detail.

SEL25 AF Mode does not make sense because that's a Live View mode which is a particularly slow way to AF. I assume you meant one of the Expanded Area AF modes but then it's possible that the AF system was locking more often that you thought but locking on to more AF-friendly parts of the scene (i.e., not the blurry dogs).

Finally, normhead is right. Without a side-by-side comparison of keeper rates from another shooter with the same lens but another camera brand, there's no way to tell if Pentax AF is ugly or beautiful.

[stillshot2]

I totally agree, I hope you don't take my report as whining, Yes I always want better but I'm simply showing what is possible or if you take it the other way what is not possible with Pentax. No hiding the bad results and showing there are still plenty of good results regardless of the bad results.

Sorry! I found your post interesting and didn't mean to imply that you in particular are whining! I was just trying to emphasize that yes, with scientific testing Pentax may be a little behind the curve, but isn't nearly as bad as many people in many threads on this site are claiming Pentax autofocus to be, backed up by my real life field testing. Just trying to stick up for the little guy here (Pentax)

[swanlefitte]

I didn't take this as whining at all. I thought it was about as objective as you could be. Heres the test, heres my measurement system and here is what i got.
Next time get a professional baseball pitcher a radar gun to track that 100mph fastball and tell me how you didn't get any focused shots in the 0.412 seconds the ball traveled. Thats whining. After all Ted Williams said he could see the laces on the ball.

[LensBeginner]

Wow... a nice test, thanks for the write-up.
The implication that skill (gasp!) matters even for autofocus is bone-chilling...

[MightyMike]

Interesting test, but......

The AF system also has an operating shutter speed that is linked to the aperture of the lens and brightness of the scene (it's in the EXIF data if you care to look). It's quite possible that AF array readings on the rapidly rippling fur, flying dogs, rapid panning, and rapid zooming were simply too blurry for the AF system to read sufficient detail.

SEL25 AF Mode does not make sense because that's a Live View mode which is a particularly slow way to AF. I assume you meant one of the Expanded Area AF modes but then it's possible that the AF system was locking more often that you thought but locking on to more AF-friendly parts of the scene (i.e., not the blurry dogs).

Finally, normhead is right. Without a side-by-side comparison of keeper rates from another shooter with the same lens but another camera brand, there's no way to tell if Pentax AF is ugly or beautiful.

In your first comment I think you're grasping, no doubt such functionality is linked but aside from the first set I was shooting wide open, and if you think zooming cause issue I actually think it did the opposite, it kept the subject a similar size in the frame making it easier for the tracking algorithms to recognize it.

as for SEL25 unless we've got a mix-up of terminology I totally don't understand how you're assuming its a LV AF mode... BTW my mistake i was using SEL33 as I just checked the camera which was left in the same settings I was shooting in. The info screen says SEL for the AF mode it is the same as Expanded Area, just acquire the subject in the center of the frame and the camera is supposed to follow it as it moves about the points, It isn't supposed to lose the subject or randomly choose let say the grass after its supposedly acquired the subject then again once started not in focus on the subject it can reacquire the subject. I suppose AF hold to medium may have been better.

There is always someone who dismisses a report because it wasn't done side by side with the competition, but heck I don't have 2 or more other clones and enough money to have matching gear on other brands to do a side by side comparison of the same subject at the same time from the same angle because you could take this requirement to the extreme, are the lenses calibrated equally? can the settings in the cameras actually be matched? how do 3 identical photographer occupy the same space at the same time? No! this is the real world! these are real world results and this is what is possible given this situation with a Pentax system and described gear! I have over a decade of experience with Pentax and host a camera club so I've used plenty of other systems however minimally. We all know that other systems are just plain and simple more advanced and regardless of these results or a side by side comparison that truth doesn't change.

As for the ugly bit in the title I think too many people are taking the word too seriously and biasing their reading of the report based on that word along, it was a bit of truth and a bit of playing on words due to the very cute subject!

People have to stop getting their panties in a knot and realize that not every report like this is meant to bash Pentax but rather to be a clear honest representation of Pentax.

---------- Post added 06-23-18 at 05:03 PM ----------

Sorry! I found your post interesting and didn't mean to imply that you in particular are whining! I was just trying to emphasize that yes, with scientific testing Pentax may be a little behind the curve, but isn't nearly as bad as many people in many threads on this site are claiming Pentax autofocus to be, backed up by my real life field testing. Just trying to stick up for the little guy here (Pentax)

Nothing wrong with that, i feel my test proves that though Pentax is behind it still very effective (that will depend on how one defines effective though)

[stevebrot]

The AF system also has an operating shutter speed that is linked to the aperture of the lens and brightness of the scene (it's in the EXIF data if you care to look)

I did but came up with nothing. I also consulted the ExifTool Pentax makernotes Tag reference with the same result.

ExifTool | Pentax Tag Reference

Copy paste of an EXIF report with the pertinent fields highlighted along with a note of what tool was used would be helpful.*


Steve

* Tag names for the makernotes section are arbitrary for any software that supports them. That is why it is helpful to identify the tool. ExifTool is fairly authoritative and exhaustive, but that doesn't mean it is without fault.

[MightyMike]

I didn't take this as whining at all. I thought it was about as objective as you could be. Heres the test, heres my measurement system and here is what i got.
Next time get a professional baseball pitcher a radar gun to track that 100mph fastball and tell me how you didn't get any focused shots in the 0.412 seconds the ball traveled. Thats whining. After all Ted Williams said he could see the laces on the ball.

LOL thats a test for whining about how much it hurts after the ball hits your gear You summary "Heres the test, heres my measurement system and here is what i got." is totally spot on

---------- Post added 06-23-18 at 05:10 PM ----------

Wow... a nice test, thanks for the write-up.
The implication that skill (gasp!) matters even for autofocus is bone-chilling...

I know you're having fun but lets expend on that. Lets take skill at a few simple levels, can you hold the center of the viewfinder on the subject? or more simply hold a telephoto lens steady?. Can you do this while following the subject and zooming at the same time for framing? Can you anticipate the subject and how the camera may behave to maximize the results? just because the camera can offload the focusing requirement from the user doesn't mean the user can offload all skill to the camera.

Glad you like the test, thanks for commenting

---------- Post added 06-23-18 at 05:11 PM ----------

I did but came up with nothing. I also consulted the ExifTool Pentax makernotes Tag reference with the same result.

ExifTool | Pentax Tag Reference

Copy paste of an EXIF report with the pertinent fields highlighted along with a note of what tool was used would be helpful.*


Steve

* Tag names for the makernotes section are arbitrary for any software that supports them. That is why it is helpful to identify the tool. ExifTool is fairly authoritative and exhaustive, but that doesn't mean it is without fault.

I'd like to see this too!