[beholder3]

But it's nice to have the K-1 in the bag for when you have a willing subject. Compared to using the same format, you just need a slightly larger camera bag to be ready. I use the same lenses on both.

That bird image is just fantastic!

[normhead]

That bird image is just fantastic!

Thanks for commenting.

It's why you lug that K-1 around with you over hill and dale, just in case. You might get a chance to use it. And to me, it's not different than changing lenses. Being able to change lenses or change bodies, gives you twice as many options as just being able to change lenses. And more options often leads to more appropriate decisions. For me, I'm going to nail a few images with the K-3, then haul the other body out of the bag and go for something special with the K-1. Much of the time you can't tell the difference. But every now and the K-1 captures one you know is above what the K-3 could have produced.

[dylansalt]

It was a few years ago...2012 I think and a Buddy and I went out to photograph wildlife. We came across a Great Horned Owl roosting in a tree and both of us proceeded to take pics of this bird.

He had a 7D and a 70-300 Canon lens, I used my K5 and 55-300. We were both checking out our monitors and comparing the pics we had got and he indicated that that my pics seemed that much sharper, colours richer, etc.

Same time, light, similar equipment...and we're both relatively competent.

Mind you I would rate the Pentax 55-300 as a far superior lens to the Canons 70-300 in terms of sharpness, color rendition & micro contrast - it also easily competes with my Nikon 70-300 VR at half its weight & size.

I only ever shot with the K10D, (which incidentally is still being used by my 14 yr old grandson) & never had a problem locking onto a subject in af - even in dim light

[Frater]

I wouldn't be obsessed too much with tracking, unless you photograph dogs running towards you, or cars or whatever else is approaching at a constant, easily predictable speed. All cameras are comparably pretty good at this easy task.
However, what makes Pentax AF often useless for any serious people photography is not related to tracking. But instead to two unrelated flaws:

1- huge AF areas, each covering a big portion of the motiv, sometimes (outside the center spot) in awkward big T or L shaped areas.
Mind that the tiny squares in the view finder are only about 1 to 10% of the true area size of the AF regions.

For the German test, where they used a two dimentional flat "wall" to approach the camera, this is very favourable for Pentax.
However in real life situation, e.g. heads of persons (or animals) being a bit further away, thus appearing small to the camera, huge AF areas tend to fail.

A Nikon D7200 would fare well, because its AF points are really only just AF points, rather than huge oddly shaped AF regions as in Pentax cameras.

2. Pentax is bad in detecting (and tracking) small, erratic, random movements, as typical for e.g portrait photography with shallow DOF.

It is a pity that the German test had only a very simplistic test setup, only simulating a flat wall coming perpendicularly towards you in absolute constant speed.

In a better setup simulating real life problems, let's expect that the K-3 would have failed a lot, confirming the bad reputation of Pentax AF implementations out in the market.

---

Last edited by Frater; 03-25-2019 at 12:23 PM.

[c.a.m]

However, what makes Pentax AF often useless for any serious people photography is not related to tracking. But instead to two unrelated flaws:

@Frater, your post introduces several interesting points, which could be even more interesting if you could indicate your personal actual experience with these AF issues.

Mind that the tiny squares in the view finder are only about 1% of the true area size of the AF regions.

I certainly agree that the true AF areas are larger than the small squares that are indicated in the viewfinder. However, it would be useful if you would indicate a reference for the "1%" value that is indicated in your post.

However in real life situation, e.g. heads of persons (or animals) being a bit further away, thus appearing small to the camera, huge AF areas tend to fail.

I think there are numerous examples of well-focused subjects of these types, including many images posted by PF members. Have you found that your Pentax camera(s) have failed in these situations?

A Nikon D7200 would fare well, because its AF points are really only just AF points

In fact, AF points are not really "points" in the true geometric sense. They are more like individual 'AF areas' and some cameras have smaller 'areas' than others. Concerning the D7200, how has that camera's AF worked for you? It would be great if you could provide a sample image or two.

Pentax is bad in detecting (and tracking) small, erratic, random movements, as typical for e.g portrait photography with shallow DOF.

Again, if you could provide some additional insight into your experiences in this area, it would reinforce this statement.

In a better setup simulating real life problems, let's expect that the K-3 would have failed a lot, confirming the bad reputation of Pentax AF implementations out in the market.

On what basis is this claim made? Without any reference to credible tests or detail of your own experiences, this statement seems to be speculation.

- Craig

[ThorSanchez]

I wouldn't be obsessed too much with tracking, unless you photograph dogs running towards you.

I took about 65 pictures of my dog chasing balls around the yard over the weekend with a K-3ii and a PLM. I have a fair amount of practice with that combination and kids playing soccer. Back button focus, AF-C, high shutter speed, high continuous shutter, TAv. But my in-focus hit rate on a running dog was about 20%. And it was noticeably worse with him running towards me.

[Frater]

I certainly agree that the true AF areas are larger than the small squares that are indicated in the viewfinder. However, it would be useful if you would indicate a reference for the "1%" value that is indicated in your post.

Agree, that 1% is a bit to harsh. The sensitive, effective AF areas are a bit more than 5 times as big as the viewfinder square, in each dimension: at least 5x5 = 25x, but the odd T and L shapes take away a bit from the sensitive AF area again, so let't settle on: between 10x and 20x mismatch between viewfinder indicated size and actual (effective) AF area size.

Whereas for Nikon, it is a pretty close match. And they don't have these odd polygonal shapes, but are straight rectangular (or cross).

I assume, that Pentax has to cheat via AF area size to maintain AF sensitivity claims, so that it doesn't fall behind Nikon in that respect too much. Wheras Nikon can do AF sensors which are sensitive naturally, so that Nikon can keep them as small as indicated in the viewfinder, and still have -3 EV or whatever.

The Problem is though, that the Pentax cheat approach causes AF to fail on threedimensional smaller targets not only in darker situations, but all light situations suffer from that, as a collateral damage.

This was for the K-5, newer models may be better or worse (if to catch up with the sensitivity claims further raised by Nikon).

But for two dimensional stuff (landscapes) (everything at infinity), the actual size of the sensitive area is irrelevant anyway Actually for landscape, you don't need AF alltogether.

---

Last edited by Frater; 03-25-2019 at 12:51 PM.

[c.a.m]

But my in-focus hit rate on a running dog was about 20%. And it was noticeably worse with him running towards me.

@ThorSanchez, I wouldn't refute your experience -- you got what you got. On the contrary, there are other accounts of better success, including at least one in this thread: How does Pentax AF-C compare to other brands? - Page 3 - PentaxForums.com

My impression is that the PLM lens (DA 55-300mm?) focuses very quickly, so you were using a fast lens in that respect.


- Craig

[c.a.m]

Agree, that 1% is a bit to harsh. The sensitive, effective AF areas are a bit more than 5 times as big as the viewfinder square, in each dimension: at least 5x5 = 25x, but the odd T and L shapes take away a bit from the sensitive AF area again, so let't settle on: between 10x and 20x mismatch between viewfinder indicated size and actual (effective) AF area size.

Thanks for your comments.

AFAIK, I don't think Ricoh/Pentax has ever published any information on the relative extent of the focus points/areas. Coincidentally, I am in the process of mapping out the AF areas on my own K-3 II. One of my objectives is to compare the various actual individual AF areas to the viewfinder squares. So far, I have discovered a variety of rectangular shapes and sizes and -- as we know -- they are larger than the VF squares. I have not determined the scale factor.

You may be interested in the following item from the Ricoh Imaging site, which describes Optimizing Focusing Accuracy with Large-Aperture Lenses:

"Highlight: The advantage of Spot setting in AF mode
The Spot setting is an autofocus mode that detects the in-focus point using a single AF sensor positioned at the center of the image field.
For PENTAX digital SLR cameras marketed after the PENTAX K-3, the Spot setting detects an area approximately 30 percent narrower than that of the Select setting, which uses a set of middle AF sensors in autofocus operation."

I am thinking that your experience or assumptions about 'Pentax AF' may be based on the K-5 or online comments from several years ago. My impression is that the AF performance of Pentax cameras and lenses seems to have improved significantly since that era, although there are few credible and detailed tests to confirm it.

Furthermore, as shown in many examples, the knowledge and proficiency of the camera's user also plays into successful focusing.

- Craig

[MarkJerling]

I took about 65 pictures of my dog chasing balls around the yard over the weekend with a K-3ii and a PLM. I have a fair amount of practice with that combination and kids playing soccer. Back button focus, AF-C, high shutter speed, high continuous shutter, TAv. But my in-focus hit rate on a running dog was about 20%. And it was noticeably worse with him running towards me.

I should send you my K-5 and cheap Sigma DG70-300mm zoom. It does substantially better with moving objects than your K-3ii.

Also, have a look at this thread: How good is Pentax AF tracking? - Page 5 - PentaxForums.com

[sutherland]

I took about 65 pictures of my dog chasing balls around the yard over the weekend with a K-3ii and a PLM. I have a fair amount of practice with that combination and kids playing soccer. Back button focus, AF-C, high shutter speed, high continuous shutter, TAv. But my in-focus hit rate on a running dog was about 20%. And it was noticeably worse with him running towards me.

Many AF System/Lens combinations have a tendency to struggle when the subject is moving towards the photographer. It's a consistent problem that is observed with any camera system on the marketplace.

I missed a series of a Gray Ghost (Northern Male Harrier) that was flying towards me with my D500 with a 500mm f/4 while using Group AF (which is a slightly larger array of center points). Blurry Harrier, but the birch tree (which was nowhere near the AF points) was sharp. The next run with a Female Harrier was far more successful:





Hit Rate is such a meaningless metric as it is case by case based on the:
- subject you are shooting
- how in-tune you are to the scene (your focus and reaction time)
- the conditions you are shooting in (lighting, atmospheric conditions)
- the settings on your camera
- etc.

Your camera/lens (we forget about the lens...but it is SO important to the situation) can only do so much and there really is no 'win':
- let it think too much, it may slow you down or focus elsewhere than intended
- let it think too little, it may search and search endlessly, never establishing focus (or quick to let it go)

You make a compromise of the combination you have to the setting you are in. There are good series and bad series. Sadly, you can't buy your way out of the unpredictability of photographing animals in an outdoor setting. It's part of the hobby.

[normhead]

2. Pentax is bad in detecting (and tracking) small, erratic, random movements, as typical for e.g portrait photography with shallow DOF.

I guess you didn't read the part of the article that said the D7200 is really poor at erratic movements with a really poor capture rate. It states, the D7200s use predictive algorithms and doesn't even confirm focus, leading to many out of focus images with erratically moving subjects. Pentax has a focus confirm that's twice as fast as a D7200, and they use it when tracking.

The test showed D7200 was only good for subjects moving at constant speeds.

I don't even use tracking for portraits. What kind of insanity would that be?

To discuss an article, you need to know what it says.

---------- Post added 03-26-19 at 11:05 AM ----------

I took about 65 pictures of my dog chasing balls around the yard over the weekend with a K-3ii and a PLM. I have a fair amount of practice with that combination and kids playing soccer. Back button focus, AF-C, high shutter speed, high continuous shutter, TAv. But my in-focus hit rate on a running dog was about 20%. And it was noticeably worse with him running towards me.

And yet I've never had a hit rate less than about 70%, and usually 100% before the dog starts to fill the frame. What is possible is what the best can do and is never defined by the worst user. The key here is, you have to wait for focus confirm on the first frame. If the camera is focused on something else when it starts tracking, it will remain focussed on that tree or whatever it is. If you didn't get a first frame focus lock tracking never works. All cameras are the same in that regard. No camera knows you want to focus on the dog without being told.

Another possibility comparing with a d7200 is that it's much better at close range. If you back yard is small, the Nikon may have a distinct advantage. Drive to park and shoot at a longer focal length and you Pentax will do fine. Again according to the article's testing, the Pentax will out perform the D7200 for the areas further away from the camera, but will be less than 50% in close. From 4 to 5 meters and beyond, your Pentax should excel.

---------- Post added 03-26-19 at 11:22 AM ----------

Taken from a sequence of my dog running to catch ball in my back yard.




Tripper says, if you like good dog pictures, change how you try and get them. Certainly don't try and do what Nikon does best with a Pentax. Do what Pentax does best.
'
Oh and incidentally, around 80% keepers on both these shoots, no misses at distance, it only missed as the dog got close to the camera. The camera won't change to suit your style. You have to change your style to suit the camera.

[Fenwoodian]

In this controlled German study, not bad, actually, a dagger to enemies of the brand without proof.

Full paper with methodology and results at: https://www.image-engineering.de/content/library/conference_papers/2017_03/a...erformance.pdf

Note the performance of the much hyped Canon 7D MkII dual pixel and Fuji XT1 systems (wasn't Fuji calling it at launch time the fastest in the world or some such?).

Example test below … they also do the same Z-axis test in low light (300lx).

.
Why should we even care about this ancient study? It was published in 2017, and the actual testing was likely done in 2016. The cameras they tested (many of which I've owned) are now 2 generations old (Fuji X-T1, Nikon D7200). However, it's sure nice to know that the current Pentax APS-C flagship (K3ii) held up well in it's day with these ancient competitors cameras. MEH...

Also, I don't understand their selection of cameras for this study. Some were Micro 4/3, and others were APS-C. Is that fair? Also, why no full frame cameras in this study?

If this study were done today pitting the Pentax K3ii vs competitors current APS-C / M4/3 flagship cameras (Fuji XT3, Nikon D500, Olympus M1X, etc.) , I expect the results would be very different.

Having said that... I have little doubt that the next Pentax APS-C flagship camera (released next year?) will be a huge improvement over the K3ii, and a camera body that will (for it's price range) be the next new leader of the pack..

In photography, gear tech is continually changing. It seems like every few months a new champion is crowned.

[sutherland]

Great shots!

it only missed as the dog got close to the camera

And that miss was probably due to the minimum focal distance of the lens.

[c.a.m]

In photography, gear tech is continually changing.

True, that!

I have little doubt that the next Pentax APS-C flagship camera (released next year?) will be a huge improvement over the K3ii, and a camera body that will (for it's price range) be the next new leader of the pack..

I think that we can realistically expect the next camera to be advanced with better performance and new features, but who knows for sure what they will be. 'Wait and see'?

It was published in 2017, and the actual testing was likely done in 2016.

According to the article, "The tests mentioned in this paper have been performed within the period of Q3/14 to Q3/16."

Why should we even care about this ancient study?

That's a broad question. I find that this type of report is informative and interesting to me, and caters to my technical/scientific background. There are few published reports on credible AF tests, so I find the few that are available to be useful as benchmarks, even if the results are several years old. I imagine that many folks don't find this technical lab-type stuff to be interesting at all.

Also, I don't understand their selection of cameras for this study. Some were Micro 4/3, and others were APS-C. Is that fair? Also, why no full frame cameras in this study?

The article doesn't explain the rationale for the selection of specific cameras. I assume their project was not intended to provide a comprehensive consumer review of the then-current camera/lens systems. I think they were interested mainly in developing test procedures and methodologies, and needed several representative systems to prove their concepts. They also seemed to be interested in drawing general conclusions about the state-of-the-art of that time. The inclusion of the K-3 II might have been coincidental.

If this study were done today pitting the Pentax K3ii vs competitors current APS-C / M4/3 flagship cameras (Fuji XT3, Nikon D500, Olympus M1X, etc.) , I expect the results would be very different.

That's an interesting expectation, which may or may not be confirmed should a similar test be conducted. Who knows? Would you have some additional insight to share on your prediction?

In this thread and in a couple of others here, I think we're finding that the scarcity of credible testing and reports makes it difficult to draw valid conclusions on the relative AF performance of camera brands and systems. We hear lots of brief anecdotes, but these typically don't remove the user from the equation. In comparing AF performance, it's essential to remove the non-system characteristics. To put it bluntly, if a person reports that they had terrible experience with Camera X's AF, was it because they had terrible technique or was it really the camera's fault?

- Craig