[FantasticMrFox]

I like you a lot, FMF ...

I like you too, oh great mate from the thread that shall not be named

... but either your technique or execution is horribly wrong.

To clarify on the issue, on the two occasions I am talking about I shot indoors (sports hall, better lit than others I've seen but still not great) with the DA 35 f/2.4, often at f/2.4 or f/2.8 to keep the shutter speed up and ISO down. Compared to the images you have posted the conditions were thus worse.

As to my settings: I kept the shutter speed at around 1/800th or 1/1000th of a second, which should rule out motion blur. I used the centre AF point because basketball players move quite erratically within a bunch of other players. The centre AF point is also considered to be the most reliable (or in case of the K3 the three vertical points). Priority for continuous shooting in AF.C was set to 'sharpness', AF hold status was set to 'medium', cont. shooting was set to high. I used back button focusing, AF.C.

If you have any recommendations on how to improve on these settings, I'd be glad to give them a try As I said, the problem didn't seem to be the focussing speed of the lens AF per se, but the fact that the camera would just let the subject move quite a bit until it would refocus. Maybe I should try 'hold status' high instead medium? I was just afraid that'd bring the already reduced frame rate down even more.

The subjects in these photos are approaching the camera at much faster than walking pace - the top one about 10-15 kmh and the bottom about 40-50 kmh (time trial by professionals).

May I ask which lenses you used and how far the subjects were away from you?

[rawr]

someone praising the higher fps in burst mode of the K3 compared to the D750 and D610

That was me. And what I intended was merely to correct norm's number in a previous post about the D750 doing 11 fps, instead of just 6.5fps.

I added the K-3 8.3 fps number because I think we often overlook what a capable machine it really is. 8 fps, with active AF, at 24MP is still very rare in a DSLR. And I believe the K-3's AF-C/tracking performance is mis-judged.

I don't want to criticize the D750. I may buy one soon if the discounts keep coming, as a K-3 'backup', when I need an extra stop or thereabouts of high-ISO IQ with decent AF, (just like I've been using the D610).

[FantasticMrFox]

I added the K-3 8.3 fps number because I think we often overlook what a capable machine it really is.

Yes, especially as DPReview actually measured it at a higher-than-stated 8.9 fps.

[normhead]

That was me. And what I intended was merely to correct norm's number in a previous post about the D750 doing 11 fps, instead of just 6.5fps.

I added the K-3 8.3 fps number because I think we often overlook what a capable machine it really is. 8 fps, with active AF, at 24MP is still very rare in a DSLR. And I believe the K-3's AF-C/tracking performance is mis-judged.

I don't want to slander the D750. I may buy one soon if the discounts keep coming, as a K-3 'backup', when I need an extra stop or thereabouts of high-ISO IQ with decent AF, (just like I've been using the D610).

Darn, why have I been thinking it was 11? I don't want one anymore....

Honestly, when we're out in the cold shooting images like these... images like these are often the best of a burst that fills the buffer. Around an image like this there may be a second on each side of frames that aren't as good. It's a mater of getting a compelling image, or settling for a very good image.



The guy with the best frame rate gets all the love in the group hug. These animals aren't posing, they're going about their business, and selecting from 8 images not 6 can make a difference. The guys with their Canon 1DX cameras and 12 to 14 fps... everyone hears those shutters going and drools. Seriously, 6 FPS is a huge drawback for wildlife, and I'm guessing it's just as big for sports. That piece of information is enough to completely change my perception of the camera. Maybe the AF isn't good on a K-3, for how some of you shoot. But for those of us used to composing for MF, the frame rate is way more important than the AF speed. AF is an option we can live without, completely.

I can tell you exactly why the Nikon and Canon guys not using high end gear ask about my K-3. They hear my shutter.

---

Last edited by normhead; 05-20-2015 at 05:03 PM.

[clackers]

As to my settings: I kept the shutter speed at around 1/800th or 1/1000th of a second, which should rule out motion blur. I used the centre AF point because basketball players move quite erratically within a bunch of other players. The centre AF point is also considered to be the most reliable (or in case of the K3 the three vertical points). Priority for continuous shooting in AF.C was set to 'sharpness', AF hold status was set to 'medium', cont. shooting was set to high. I used back button focusing, AF.C. .... May I ask which lenses you used and how far the subjects were away from you?

Yeah, if you can isolate the moving subject, you can maximise the tracking by selecting a large expansion area.


You're going to get limbs and bodies thrown in front of your basketball player, though, so this is challenging for any camera. If you're posted at one end, you can switch to MF and pre-focus on the distance they shoot from. I do MF with bird photography wherever there are lots of distracting branches and leaves present.


Your shutter speed's fine for the top of jumps and so on. For the sideways running, go to even 1/1500s if you want to freeze the action.


Since you're using back button focusing (I really should do this, too), turn the AF hold status to 'Off'.


You should close down that aperture, though ... f2.8 from relatively close up (you're using just a 35mm lens) will result in most things being out of focus. The depth of field should also cover a nearby team mate lending support or an opponent doing a challenge, unless you wanted an individual portrait.


I used a Tamron 70-200mm in both cases.


The lens was closed down to f4, but DoF is fine because of the distance to the subject - maybe 80 metres from memory in the mother/daughter example, maybe 30 metres for the Sky rider.

[Christine Tham]

The guys with their Canon 1DX cameras and 12 to 14 fps... everyone hears those shutters going and drools.

Technique and timing matters more than frame rate.

I remember doing some shots at a triathlon with my K-5 and I was getting better results than a friend with a 1D. Then we swapped cameras, because he was seriously starting to think the K-5 must be a killer camera, and I was taking even better shots with the 1D. And he was seriously pathetic on the K-5, because his technique amounted to "press the shutter, and hold it down until the buffer runs out"

The 1D series has an amazingly deep buffer, but the K-5 was pretty good (particularly when shooting JPEG).

Tracking is overrated. If the subject is coming towards you, you are better off using MF because the brain is a much more sophisticated predictor than what a puny processor on a camera can do. Photographers were capturing sports and wildlife long before AF was invented.

[Ian Stuart Forsyth]

Which is totally irrelevant when 80% of the K3 shots are out of focus due to far inferior AF.C and slow focusing motors in all Pentax lenses, except the new 70-200 and 150-450 maybe.

You have a valid point, for action photography I would pick keeping the frame rate at 6 FPS if the trade-off was reliable AFC tracking with accurate target recognition, the additional 2-3 FPS won’t make that much of a difference if AFC is less than ideal. Playing around with the higher end C &N cameras in 2012 was kind of an eye opener to what they have accomplished with regard to their AF performance.

While it’s not fair to compare a camera that is $1000 more and expect them to have the same AF performance I would sooner spend the additional and get the performance I want. Buy the camera with the performance I want and replace it when its kicked the bucket, This way I am not left buy the next generation of camera with my fingers crossed hoping for better performance. When we consider the cost of the K5II and the upgraded to a K3 it puts you in the ballpark of a higher end DLSR any how.

[Rondec]

You have a valid point, for action photography I would pick keeping the frame rate at 6 FPS if the trade-off was reliable AFC tracking with accurate target recognition, the additional 2-3 FPS won’t make that much of a difference if AFC is less than ideal. Playing around with the higher end C &N cameras in 2012 was kind of an eye opener to what they have accomplished with regard to their AF performance.

While it’s not fair to compare a camera that is $1000 more and expect them to have the same AF performance I would sooner spend the additional and get the performance I want. Buy the camera with the performance I want and replace it when its kicked the bucket, This way I am not left buy the next generation of camera with my fingers crossed hoping for better performance. When we consider the cost of the K5II and the upgraded to a K3 it puts you in the ballpark of a higher end DLSR any how.

We'll see. The K3 II supposedly has improved auto focus capability and it will be released at 1100, meaning that street price will be in the 900 dollar range in a few months. K3 is currently selling in the 700 dollar range. Nikon D750 is 2000 (body only) -- B and H prices.

Honestly, low light situations with poor contrast are a miserable shooting experience with any camera. And just because the camera appears to lock doesn't mean that it actually has focused correctly. I don't shoot in those sorts of situations a lot, as I don't find the results to be great, even if the focus is spot on, but I am not a concert shooter either.

[normhead]

Technique and timing matters more than frame rate.

Except for the part where you can have technique, timing and frame rate. I'm not sure why you'd frame a discussion as if having one means you don't have the other.
or to state things simply...

If technique and timing are the same, the guy with the highest frame rate will have the best chance of coming away with the best image. At least in wildlife and sports and anything else that involves action.

I know why some of my buddies shoot with Canon 1Dx cameras, I just can't afford one.

When I was 20 years old, my reaction times was about .08 seconds. Now its about .25 seconds. And that doesn't take into account shutter lag. SO when I was 20 shooting what I see would have always meant .08 seconds late, bad but not real bad. Now it's .25 seconds late, real bad... everyone I shoot with starts their shutter in anticipation of a moment and depends on frame rate to get the image we want. it's pretty funny when the shutters are all going and the animal doesn't make the anticipated move, and you end up with 23 useless frames. The language gets a little colourful, and sometimes the animal does what you were waiting for while you are clearing your buffer. But even in that case, you can still get off a few frames.

If you suck, maybe frame rate doesn't help you. You just get lots of bad pictures. if you know what you're doing, it's a wonderful thing. You get to select from a burst of pictures any one of which is as good as anyone else's single shot, but you have more chance to get the exact moment when photographically it all comes together instead of the a split second before or after. You can (theoretically) get the best shot without frame rate, but a fast frame rate definitely increases your odds.

I tried some action shots using burst with my K20D the other day, after using a K-3 it's just totally depressing... I'll never use it that way again..

Buying a camera for action shots with a lower frame rate than a K-3 just doesn't make any sense to me.

I don't want to trade "this for that". I want better everywhere, if I'm going to switch, not trading one problem for another.

[rawr]

if I'm going to switch, not trading one problem for another.

If you really need fps, Sony A6000 does 11 fps with 179 focus points, Samsung NX-1 does 15 fps with 209 focus points. Both much cheaper than a 1Dx.

[normhead]

If you really need fps, Sony A6000 does 11 fps with 179 focus points, Samsung NX-1 does 15 fps with 209 focus points. Both much cheaper than a 1Dx.

And everything else is equal or better?

Shot outdoors with my K-3 a few minutes ago..... AFC, focus priority, hold level set to 4 , 14 shots in 2 seconds, every shot in acceptable focus, dog trotting at an even pace...

14 shot sequence...


First image...


Last image...


First and last uncropped.







Really guys, I don't know if you're trolling, or you honestly don't know or don't care what the camera is capable of. Clear signs of focus shifting (tracking) through the 2 seconds. Maybe other systems are better, but really, so what?

I really hate it when I test these bogus theories , and find something different than what's been posted. You can't beat this with a D610, a D810, A D750 a Canon 6D, most of the cameras on the market, because you can't match the frame rate, and as long as the K-3 can keep up, as in this trotting dog, it's simply better. Honest to god, I spend more time in here investigating BS that just doesn't sound right, and isn't.

Or more to the point, just because you or some reviewer can't figure something out, doesn't mean someone else can't.

Actually going out and doing it, trumps all reviews. That's how you find out what your camera is capable of.

There are ample examples on the forum of both Pentax AF and tracking, this crap is lazy, un-informative and lacks context. So, I've established that for 14 exposures over 2 seconds, the k-3 is adequate, and that you can't get 14 exposure on most cameras over the same time frame, even if they can match the AF. Now you guys with all the other systems you're touting show me when the system you think is so hot is better. Not what some idiot reviewer says... show me, like I showed you.

Everyone knows every friggin thing about why everything else is better than a Pentax, but not one of you seems to know what a Pentax can do. But you do know that some half qualified reviewer, says some camera some freakin where in the world can do whatever it is, better. than a Pentax can. And you buy into that nonsense.

Let me play that game for a second.. if you're shooting a black and white border collie coming at you at a mild trot, you can capture 7 frames a second on a K-3 better than you can on a D750, a 6D , a D810, a D610 and a pile of other cameras that cost way more than it does... and that is an absolute fact. And whatever else is true or not true, those cameras, in this one aspect, are inferior to a K-3. Now then, what is it you think you know?

[rawr]

Now then, what is it you think you know?

I hope you don't think that I am the one arguing with you.

I agree that the K-3 AF works really well, including tracking. Combine that with 8 'active AF' fps and it's all impressive, and it shows the K-3 punching above it's weight.

AF tracking with the K-3 (or any other camera) is one of those craft skills, like panning with slow shutter speeds during action sports, that needs practice to develop. That's the failing of a lot of the Youtube testers of AF-C tracking with the K-3. They play with a few cameras for 30 minutes over some dumb scene, with a slow lens, and then draw conclusions about the K-3 relative to others that aren't really valid.

I've yet to see a tester, for example, use the same lens - eg the Sigma 70-200 f2.8 OS, which has fast AF and is available for Pentax, Canon, Nikon and even Sony I think - on similar K-3 level cameras (eg Nikon D7200, not a Nikon D4) in the hands of an experienced action photog to test AF-C tracking performance.

It took me a while to get comfortable with the K-3's AF-C, and I lost lots of shots due to that. But now I feel more confident with it and can get better results. Ditto with Nikon's '3D-Tracking' AF-C on a D610 - I couldn't figure out how it was meant to work at all for several shoots, and so I have thousands of out-of-focus Nikon photos which I could easily use to [wrongly] claim that Nikon AF-C tracking on FF is garbage.

[normhead]

I hope you don't think that I am the one arguing with you.

I agree that the K-3 AF works really well, including tracking. Combine that with 8 'active AF' fps and it's all impressive, and it shows the K-3 punching above it's weight.

AF tracking with the K-3 (or any other camera) is one of those craft skills, like panning with slow shutter speeds during action sports, that needs practice to develop. That's the failing of a lot of the Youtube testers of AF-C tracking with the K-3. They play with a few cameras for 30 minutes over some dumb scene, with a slow lens, and then draw conclusions about the K-3 relative to others that aren't really valid.

I've yet to see a tester, for example, use the same lens - eg the Sigma 70-200 f2.8 OS, which has fast AF and is available for Pentax, Canon, Nikon and even Sony I think - on similar K-3 level cameras (eg Nikon D7200, not a Nikon D4) in the hands of an experienced action photog to test AF-C tracking performance.

It took me a while to get comfortable with the K-3's AF-C, and I lost lots of shots due to that. But now I feel more confident with it and can get better results. Ditto with Nikon's '3D-Tracking' AF-C on a D610 - I couldn't figure out how it was meant to work at all for several shoots, and so I have thousands of out-of-focus Nikon photos which I could easily use to [wrongly] claim that Nikon AF-C tracking on FF is garbage.

And that is the approach of someone who's actually learning to use his gear, as opposed to writing a quick review. The funniest thing about this too me, is you can tell from what rawr wrote, he's spent some time figuring out the cameras... and you can tell by reading what many reviewers have written, they didn't.

[aleonx3]

And that is the approach of someone who's actually learning to use his gear, as opposed to writing a quick review. The funniest thing about this too me, is you can tell from what rawr wrote, he's spent some time figuring out the cameras... and you can tell by reading what many reviewers have written, they didn't.

I am glad I chime in to this thread... learn a great deal from the posters here... but I still don't like the D750 (it is in my hands right now); OTOH if k-3 doesn't not exist then this would probably be the best for me anyway.

[Ian Stuart Forsyth]

And everything else is equal or better?

Shot outdoors with my K-3 a few minutes ago..... AFC, focus priority, hold level set to 4 , 14 shots in 2 seconds, every shot in acceptable focus, dog trotting at an even pace...

14 shot sequence...


First image...


Last image...


First and last uncropped.







Really guys, I don't know if you're trolling, or you honestly don't know or don't care what the camera is capable of. Clear signs of focus shifting (tracking) through the 2 seconds. Maybe other systems are better, but really, so what?

I really hate it when I test these bogus theories , and find something different than what's been posted. You can't beat this with a D610, a D810, A D750 a Canon 6D, most of the cameras on the market, because you can't match the frame rate, and as long as the K-3 can keep up, as in this trotting dog, it's simply better. Honest to god, I spend more time in here investigating BS that just doesn't sound right, and isn't.

Or more to the point, just because you or some reviewer can't figure something out, doesn't mean someone else can't.

Actually going out and doing it, trumps all reviews. That's how you find out what your camera is capable of.

There are ample examples on the forum of both Pentax AF and tracking, this crap is lazy, un-informative and lacks context. So, I've established that for 14 exposures over 2 seconds, the k-3 is adequate, and that you can't get 14 exposure on most cameras over the same time frame, even if they can match the AF. Now you guys with all the other systems you're touting show me when the system you think is so hot is better. Not what some idiot reviewer says... show me, like I showed you.

What this series of photographs tell me is that in bright light at F8 with a 200mm lens that the K3 still struggles with AFC tracking and keeping it focus, just as when I tested it with the sigma 300 2.8 and the hsm 70-200 F2.8