[ThorSanchez]

What do you want, actual research or a few generalized assumptions.
Personally, I don't give in to anyone's generalized assumptions. That's not scientific. Science is there to test generalized assumptions, and it'a amazing how often they are wrong.

It is quite reasonable to assume there will be some difference. Whether or not it's a significant difference requires research, how much the proper settings would make is also unknown. So no, it's not reasonable to consider "generalized opinions" If we did that we'd all just buy A9s 1Dxs or D4's and our photography for 99% of us would suffer greatly.

Given that there is little or no solid, applicable, up-to-date research with good starting conditions and use cases I'll use the best available data, which includes a good amount of subjective information.

When there is a large body of mostly subjective review data available, and one somewhat questionable 5+ year old semi-objective study you have to weight those things appropriately. I'm not ignoring the results of 10 double-blind, comprehensive, controlled autofocus studies. That simply doesn't exist. I'm looking at a reasonable consensus of the available information.

If you choose to dismiss anything that isn't conducted to rigorous scientific standards that's your choice. But you'll be making decisions based on a tiny fraction of the available, relevant information.

[Ian Stuart Forsyth]

What would be nice would be if a few of the folks who claimed the tests posted above were inaccurate would do some comparisons,

There has been people who have posted such test. and even when I use a third party lens and use it well outside of what I would normally think of using the lens for I find that I cannot replicate their lack of keepers.
Sigma 150-600 sport vs $4400-$8000 lenses - Page 2 - PentaxForums.com
Here I am using the sigma 150-600 sport with a 1.4tc, this is a combo that I would not recommended ( and any person that uses Nikon for action would also not recommend ) for action work but I had very few images that the camera and lens could not track.

Here is the Nikon 70-200 F4 something I would not consider as a action lens but to get the lens/d800 to fail was very infrequent while tracking moving targets traveling at 110 kph

here is the final image from the sequence where the truck is somewhere in between 15-7 m away as recorded in the camera data.
as you can see from above where I focused and below you can see how much the camera missed with its AF

here is a 100% crop
https://photos.smugmug.com/Temp/Temp/i-xbdggpk/0/d9fd2ca5/O/missed.jpg
Here is the entire sequence as the truck moved towards me
isfphotography Photo Keywords: action

even at the time where there is the most difficulty in predicting where to focus you can see that the camera/lens combo had really no trouble it capturing the subject. For me when I first started testing the camera it was mind boggling what would be needed to calculate this accuracy in predicting the timing needed to achieve this, where the vehicle is traveling at 30 meters per second and placing the focus plane within 0.15 meters of the targeted subject and when all things are consider it only takes less than a 100th of a second for the target to travel that 0.15m distance.

Now I never really worry as to how this is can be done and focus on that it can be done.

[Dan Rentea]

There has been people who have posted such test. and even when I use a third party lens and use it well outside of what I would normally think of using the lens for I find that I cannot replicate their lack of keepers.
Sigma 150-600 sport vs $4400-$8000 lenses - Page 2 - PentaxForums.com
Here I am using the sigma 150-600 sport with a 1.4tc, this is a combo that I would not recommended ( and any person that uses Nikon for action would also not recommend ) for action work but I had very few images that the camera and lens could not track.

Here is the Nikon 70-200 F4 something I would not consider as a action lens but to get the lens/d800 to fail was very infrequent while tracking moving targets traveling at 110 kph

here is the final image from the sequence where the truck is somewhere in between 15-7 m away as recorded in the camera data.
as you can see from above where I focused and below you can see how much the camera missed with its AF

here is a 100% crop
https://photos.smugmug.com/Temp/Temp/i-xbdggpk/0/d9fd2ca5/O/missed.jpg
Here is the entire sequence as the truck moved towards me
isfphotography Photo Keywords: action

even at the time where there is the most difficulty in predicting where to focus you can see that the camera/lens combo had really no trouble it capturing the subject. For me when I first started testing the camera it was mind boggling what would be needed to calculate this accuracy in predicting the timing needed to achieve this, where the vehicle is traveling at 30 meters per second and placing the focus plane within 0.15 meters of the targeted subject and when all things are consider it only takes less than a 100th of a second for the target to travel that 0.15m distance.

Now I never really worry as to how this is can be done and focus on that it can be done.

You know what else it's questionable in the german lab test, besides the wrong Af settings?

They tested the performance of the cameras at an aperture of f5.6 and the ISO speed was set so that they have and I quote a no longer exposure time than 1/200s for any condition. Also, for the published results they used a speed of 0.4m/s (a walking person) and 1m/s (a running person).

What's the DOF for a 50mm lens used on a D810 like they used in the test and shown in the print screen who made Normhead to start such a debate? What's the recommended shutter speed to freeze a running person, especially when we talk about a 36mp camera also? Or for a 50mp (Caon 5DsR) camera they also tested?

The Cambridge DOF calculator shows the DOF on a full frame at f5.6 and a distance of 5m (see the image below). Yet, they managed to get 33% keeper rate with D810 and a 50mm lens on a subject that raced for 3m. The subject started "the race" from a distance of 5m from the camera and finished "the race" at a distance of 2m from the camera while running at 1m/s (that's 100s for each 100m; knowing that the best athletes will run 100m in 9.8 seconds, this means that in the german lab test the person was not running fast at all; I would say a relaxed warming jog rather than running).

As I said, even without the wrong Af settings this test is questionable knowing the other parameters involved in the test.

[Dan Rentea]

Here, another test, this time it was taken with a 35mm f2 lens (yes, you read correctly, a 35mm lens for action ) on a full frame camera. Real world tests with a bird that is flying with a lot more than 1m/s. Given the fact that even in Monte Carlo I wasn't interested in landscapes, I saw this seagull circling the area and I thought, why not take some shots of it for testing af with a wide angle lens, which is challenging for various reasons?!

Below it's a series of 7 images straight out of camera from a burst taken with a camera that has 7fps (5D Mark IV) so it's a burst of one second. Settings were 1/2500s, f5.6, ISO set to Auto. The focus could have jumped right away to the crane seen in the background because it's bigger than the seagull, it's on the route of the seagull also and it's more contrasting so easier to fool the af system...

Image No. 1



Print screen of the Image No. 1 zoomed at 100% (you can see in the left corner of Photoshop window that is zoomed at 100%)



Image No. 2



Print screen of the Image No. 2 zoomed at 100%



Image No. 3



Print screen of the Image No. 3 zoomed at 100%



Image No. 4



Print screen of the Image No. 4 zoomed at 100%



Image No. 5



Print screen of the Image No. 5 zoomed at 100%



Image No. 6



Print screen of the Image No. 6 zoomed at 100%



Image No. 7



Print screen of the Image No. 7 zoomed at 100%



And below it's the same bird photographed with the same camera, lens and settings the second time she flew towards me. I went for one shot following the bird in the viewfinder until she came close enough to me to take the shot and fill the frame with the bird.



I tried so hard yesterday to find on my hard drives bursts of images with so bad keeper rate as the german lab results were and I gave up because as I said, it's hard to beat those poor results shooting in real life conditions. So, if people ignore all the real life images posted by me and Ian Stuart Forsyth and make a case with a lab test, then photography world gone completely mad and now I'm not wondering anymore why DPreview and Youtube influencers gain so much popularity with their lab tests or paid reviews...

Next time someone asks about a good camera for action we should recommend Leica-T because it blows competition in terms of tracking with it's contrast detect af. And we can't be blamed for our recommendation either because we have a german lab print screen as an irrefutably argument.

---

Last edited by Dan Rentea; 03-21-2020 at 06:08 AM.

[clackers]

I think, Dan and Ian, you are completely missing the point of the scientific method.

You are suffering from different photography skills and different scenes and different subjects and different methodologies. Dan could pull out his Canon images, Ian his Nikon, and me my Pentax, but they're *not* the same how much you claim.

The idea is standardized, reproducible circumstances and methodology that are exactly the same between each test and each brand - this is what the German lab did and both of you have failed to do!

What you are doing is not valid comparison, but attempting promotion of the brands you've bought into, in a Pentax forum. That's - impolite. Forbidden elsewhere, such as in the Pentaxians Facebook group, that's not just unwelcome, it's grounds for banning.

[Ian Stuart Forsyth]

I think, Dan and Ian, you are completely missing the point of the scientific method.

No I completely get the point, we are here to take with undying faith of a test ( for Nikon you can not even set the setting to how they have published their test) and believe it as fact and not be given into questioning such testing.

These test don't reflect what others have been doing without any problems for 1/2 a decade and we are to believe them and take them as gospel.

When we point out that this testing has no real bearing in real life uses its impolite and that we should be banned for this, when this day comes the forum will cease to a forum and will be lowered to a fanboy blog with no critical thinking as to what Pentax is.

If the day ever comes that Pentax did reach the level of AF performance as other companies we would see a great deal of threads about how much of an improvement this next generation of camera has over the previous model.

We see it every time a new model comes out " wow its soooooo much bettttteeeerrrr than the last model" but this time we would see the improvement that we see from other companies their should be a great deal of people eating crow as to how bad the older models are. And should be banned from buying such camera

One of the problems is you cannot sanitise down AF into a very simple test, it only really comes from repeated use over time as to come to a conclusion over an array of different uses.

Here is a very simple AF test anyone can do, take a photograph of a subject that is off in the distance and as quickly as you can take a photograph of a target with a slant edge near MFD, give us the raw file or out of camera jpg so that we can see the time and the accuracy of the camera.

This should not be difficult for someone like you to provide. I will do the same test and then we have some data to compare. With Pentax their is not a camera that I have had access to that can come close to what I can with another brand of camera, even with a tc added to combo the outcome remains the same.

I am wait for you to publish your results ? you can do so in the Non-Pentax forum as not to contaminate where this maybe offensive to some users.

I always welcome these discussions as they shed light into issues I observe and would rather discuss them and test them without the sanitisation of Fanboyism unless this is what you want.

Not saying this is you but their are a few that would push this discussion until things get banned, deleted or closed. I would rather that the discussion be allowed to continue and to deal with the individuals and to simple be blocked from the discussion.

Million dollar question would be, can this happen

Sadly here I don't think it would happen. ¯\_(ツ)_/¯

[mikesbike]

Yet another thread turned into a very lengthy and arduous treatise to persuade how superior Nikon and Canon are over Pentax if one wishes the best results in photography. Apparently, not even is Pentax good enough for soccer dads and moms! (this thread) Even though some good indoor basketball shots using Pentax equipment were offered, and not even the best possible Pentax equipment! Last time such an effort was made to persuade that Nikon or Canon are the answer, it was about wildlife photography. Now this.

It seems nearly every thread coming in here should add an auto-prefix to the title- "Why should I not switch to Nikon or Canon because:" then comes the issue. We might as well become a running advertisement for Nikon and Canon. After all, any test done showing results indicating otherwise is absolute crap. Videos of other photographers showing very good AF results from Pentax are flawed. All photos of like kind taken with Pentax are questionable. Only these photos taken using Nikon and Canon are proof-
of their superiority. And of course, Canon or Nikon forums are certainly never "fanboy blogs".

(And BTW, Nikon and Canon come out with new models that are "much better" far more often than does Pentax.)

That many of us here have had good success using Pentax equipment for action shots is, of course, highly questionable.

---

Last edited by mikesbike; 03-23-2020 at 01:32 AM.

[ThorSanchez]

That many of us here have had good success using Pentax equipment for action shots is, of course, highly questionable.

It is of course true that you or I can get good action photographs with a Pentax camera, while other brands have more advanced autofocus systems. You can get some good action shots with a purely manual focus camera. And it is possible to point this out without being a traitor to the Pentax cause, nor advocating that people abandon the brand en masse.


I picked Pentax for a number of reasons. Great value for the money. Lots of features that other brands reserve for much more expensive cameras. Weather sealing. Rugged, well-built bodies. Awesome little prime lenses. The K-30 came in this cool shade of blue. And on and on. Those are all still valid reasons. Discussing the autofocus performance of the brand in comparison to other brands and suggesting that perhaps Pentax could do better in future bodies does not make me anti-Pentax, or an advocate for some other brand. I certainly hope it doesn't get me banned.

Every brand has strengths and weaknesses. I would hope we're not so thin-skinned and hypersensitive that the specter of banning has to come up every time we talk about things Pentax cameras may not do quite so well, and hope they improve upon in the future.

[clackers]

No I completely get the point, we are here to take with undying faith of a test ( for Nikon you can not even set the setting to how they have published their test) and believe it as fact and not be given into questioning such testing.

These test don't reflect what others have been doing without any problems for 1/2 a decade and we are to believe them and take them as gospel.

Ian, the scientific method says you need to provide better data with better methodology to overturn their findings. Until then, it's just you puffing your chest out. And we don't take that seriously, that's not comparative data.

That's going to be difficult. You and Dan are not in that position, you're not a professional testing facility, you just do what the rest of us do, shoot subjects in our own style with our own equipment.

Here is a very simple AF test anyone can do, take a photograph of a subject that is off in the distance and as quickly as you can take a photograph of a target with a slant edge near MFD, give us the raw file or out of camera jpg so that we can see the time and the accuracy of the camera.

This should not be difficult for someone like you to provide. I will do the same test and then we have some data to compare.

Genuinely, this is nuts. Unless you make it a job, hire me for a minimum one day shoot (8 hours) at $125 per hour, I'm not interested. I will take PayPal, bank transfer or Bitcoin, BTW!

You can hire the Germans too, I'm sure, with all their equipment, to do a similar experiment in their lab with all their different brands of camera, and come up with methodology acceptable to them.

If you like, Image Resource already does static AF timing tests as part of their reviews, but of course, they're only semi-serious about them, they're a tiny internet review website and don't claim to be anything else, they're not engineers: Pentax K-1 II Review - Performance

And Ian, you'll get a lot more sympathy to your prejudices at the sister site Nikon Forums, not here. I can't speak for how worthwhile the experience is, or even if many people use it. Same with Canon Forums, it doesn't match here, either, even though both have been running for years now, not months. Perhaps Pentaxians are more interested and interesting people.

Us Pentax owners are here on a forum I've helped pay to keep running over the years, you have't. What's a word for someone who won't contribute to the costs of a community? I'll let you come up with one.

If you want to deliver 'full and frank' backhanded insults to the brand and its users, DPR is the traditional home of all that!

[mikesbike]

@ThorSanchez- Of course. And it is common here to advise new inquirers as to choosing Pentax- first question to them is what is their photographic interests. It is then common to advise that AF tracking for fast sports action, etc where extensive burst shooting is a frequent need, as well as the best results for video, Pentax might not be the best solution. Then come the strengths Pentax offers, and that these concerns mentioned do not mean Pentax cannot handle such needs.

OTOH, here in this thread we have the OP who already owns Pentax equipment and is looking for improved performance, perhaps related to technique. The inquiry is with a budget of around $500 what might that buy for better results. This is a different story altogether. And the experience of many users, as well as some tests have indicated there is not such a huge difference in AF with current Pentax equipment compared to others.

Just yesterday, I was at a park by the water for some exercise. I did not have a camera with me. I returned to the car, parked facing the water, listening to music as darkness fell. There were other vehicles there also, and a couple more cars pulled in. A young man in a suit along with several other people got out and went down to the water's edge. One young lady had a DSLR and was taking photos of the young man, positioned with the large, brightly-lit bridge to Canada at a distance for background. I paid little attention until I realized they were returning to their vehicles, then going out again for another try, and that no flash was being employed. It was evident they were having problems getting a satisfactory photo. So I walked over to help if I could.

There were streetlights along the walkway by the water's edge, so I could see that a smaller Nikon was being used, a D3400 I believe, along with some sort of kit lens. The young lady knew nothing about using it. The camera kept adjusting for the background as best it could and doing a fair job, but leaving the young man in the dark. If she popped up the flash, it would set a flash sync speed too fast so the background went dark. I asked if it had a Scene mode for night shots with flash- she didn't know. Others in the party were getting impatient to leave. It was his HS graduation. Not having much time, I advised shooting in Manual mode and ISO 3200, aperture f/5.6 which should provide decent DOF at the distance of the subject and for the bridge, with a shutter speed of 1/30 sec. I watched while she fiddled around. She didn't know how to set both, the camera not having front and rear dials. The camera was hanging around her neck, but I was able to get hold of it enough to show her how to hold down the +/- button to make the one dial work for dual-purpose. So I managed to get the aperture and shutter speed set. She took a shot but said the flash made him too bright. With the others beckoning to leave, I just said, ok walk a bit farther away and the flash will be weaker, which she did, and then she loved the result. I thought it could use just a bit more exposure, and never found out if she actually set ISO 3200, but then I thought they are all going to look at the shot with their phones, a backlit screen anyway. And precipitation was about to begin, so they were leaving. The mother thanked me for the help.

Of course it never occurred to me to tell them they should consider getting a Pentax DSLR instead, which has great low-light capabilities, and has WR construction, as well as a 2-dial system for controls- blah, blah blah. I did tell them that their camera no doubt has a control for flash output, and getting a manual for it perhaps by download, would be a good idea.

[Ian Stuart Forsyth]

Ian, the scientific method says you need to provide better data with better methodology to overturn their findings. Until then, it's just you puffing your chest out. And we don't take that seriously, that's not comparative data.

Funny you should say this, how can I provide any better data as the very data they have published there is no in setting up the camera as to how they have described setting the camera for there testing? I don't have to puff my chest all I have to do is pointing this out and it really should tell the reader maybe they should question to as what are they really testing?

that's not comparative data.
That's going to be difficult.

I don't have to be a professional tester to just look at my camera, see that there is no way to test a camera with a setting they described, that cannot be found within the limits of how you can setup the cameras AF system.

Us Pentax owners are here on a forum I've helped pay to keep running over the years, you have't. What's a word for someone who won't contribute to the costs of a community?

I have early when I first started here, besides there are many way to contribute to a community than a dollar amount. Take several post I make here are link off of this site directing traffic to this site, giving honest feed back to users with problems such as this very thread and point out that findings from other sites to reflect my on use of the equipment. I also gave away custom profiles to users here for any camera's that I had profiled, I no longer do so around here as there is now a feeling like Goodfellas club here more than ever.

The funny thing you say us Pentax owners as I have pentax cameras and several so should my voices weigh any different to the other Pentax owners just because if find another brand of camera better for some applications.

If you want to deliver 'full and frank' backhanded insults to the brand and its users, DPR is the traditional home of all that!

Funny how pointing out that my own opinions differ from that of yours and they very testing review of other brands you brought into the discussion as a backhanded insult to a brand and its users, this is why you can never have a civil conversation about the limitations of a product here. It now becomes personal to the user of that equipment for GOD sakes its a camera and not one of your children, your wife or your sister.

[clackers]

Funny you should say this, how can I provide any better data as the very data they have published there is no in setting up the camera as to how they have described setting the camera for there testing?

What are you talking about? They use ISO standard 15781 for autofocus for the testing, you never have. They used ISO standard 12233:2014 for resolution methods so it's objective MTF measurements instead of you claiming this one's in focus, this one's not. They repeated tests on 320 camera/lens combinations, you didn't. p224 says all cameras were setup by the manufacturers' recommendations in their manuals for tracking moving objects.

You're free to take up Lens Rentals' testing procedure with Roger Cicala, or Uwe Artmann of this paper ... but if you're just a random nuisance from the internet, good luck to you. In my own line of work, I get unsolicited crappy advice from amateurs all the time. You can't be crying into a handkerchief to them, saying you thought your Nikon would do better against other brands, they'll sniff your partisanship straight away. It's not academic.

So I'll say it again, come up with your own accurate, reproducible comparative data with internationally recognized standards, and present it to us and them, that's how science works.

I have early when I first started here, besides there are many way to contribute to a community than a dollar amount.

Ian, it would be parasitic to think that your personal opinions offset the money needed to run this site. They don't. If I do both, you can. In fact, it could be said that a Chevrolet owner sitting at the back of a Ford Mustang owners meeting and butting in is ... ridiculous. That would be socially unacceptable, and your only excuse would be was that it was happening over the internet, not face to face. Ridiculous again.

The funny thing you say us Pentax owners as I have pentax cameras and

Frankly, that's a terrible line for you to pursue. I also own and shoot Sony, Nikon and Samsung cameras and own four EF mount lenses. But unlike you, I politely remember what forum I'm in, who's house I'm in. DPR is the place you're looking for, to engage in disguised or quite open brand bashing of each other, they love it, it's a sick bearpit. I bet you're already there. It would figure!

[TerryL]

Totally amateur photographer here shooting my kids basketball and soccer games. I typically distribute the e-photos to families throughout the season but nothing that gets printed or anything.

I shoot with a K30 and a sigma 70-200 lens and a tamron 17-50 lens. I am considering upgrading to something better but not sure how much improvement I'd get with something else - I don't consider myself a skilled photographer. I read another forum that recommended the K-3 II for sports photography. But not sure how much of an upgrade that would be over the K30 or maybe the K-3 would be almost as good at a cheaper price. The idea of having two camera bodies by my side is appealing though - the K30 with the 17-50 and the Kxx with the 70-200.

I'd like to stick below $500 for the body and buying used is fine with me.

Thanks for any input.

There’s a
K3II in the Marketplace now. I would offer $500.

[angerdan]

Here is just one set of *objective* results with a reproducible method from a German lab. Incredible, isn't it?

From which testlab is this chart and in which magazine has it been published?

[clackers]

From which testlab is this chart and in which magazine has it been published?

Enjoy!

Auto Focus Performance ?What can we expect from todays cameras?