[janneman]

I did a lot of low light shooting at carnaval... This eyear and any other year...
The succesrate of my K20D and K-7 against for example Canon and Nikon was, well, overwhelming.. No Nikon or Canon shooter got the images I got.


Now here is the reason why.
I did not see ANY other shooter at night, in the snow and freezing cold... No Canon, Nikon, Olympus, Sony, Fuji, Sigma shooter dared to take out the camera in rain, snow, beer, chips, mayonaise environment....

Bottomline: superior AF has no use at all if you don't use it...

[Jonson PL]

Thanks for the link.
The AF tracking of the K7 should be acceptable, but the frame rate drops quite a bit.

One user left Pentax to pursue his interest in motorsport, but he commented that his first outings, with the D300 as I recall, didn't give him more keepers. But of cause there is a learning curve

[jonlee]

Thanks for the link.
The AF tracking of the K7 should be acceptable, but the frame rate drops quite a bit.

One user left Pentax to pursue his interest in motorsport, but he commented that his first outings, with the D300 as I recall, didn't give him more keepers. But of cause there is a learning curve

Not sure what would cause the frame rate to drop on the K-7. It should continue to fire even if the subject is out of focus.

AF tracking on the D300/D700 takes some time to master. There are many (too many???) settings available. While in continuous AF mode, you can track subjects based on 1, 9, 21, or 51 AF points. If using 51 points, you can also add-in "3D" tracking to the mix. On top of all that, you can have the shutter release only if the camera achieves focus, regardless of focus, or somewhere in between. You can also set the duration of how long something remains in focus before it switches to lock on to something passing in between you and what was previously in focus. My point is that it takes some practice to get the most out of any camera, given a specific scenario.

Somewhat off-topic, but is there any chance a K-x may actually track better than a K-7? I noticed the K-x has 2 AF modes - AF1 turns on 5 AF points and AF2 activates 11 (I think) AF points. Is it possible that the K-x could track smaller subjects faster using the smaller grouping of 5 points, instead of 11? Has anyone commented on the K-x's ability to track?

[mawz]

The sigma I was using was a 70-200mm 2.8 hsm.

Which version? The first two versions of the 70-200 HSM are known to be poor performers in terms of focus speed, especially the original (which has an aperture ring)

[pentaxdude]

Which version? The first two versions of the 70-200 HSM are known to be poor performers in terms of focus speed, especially the original (which has an aperture ring)

This is the first version. It has the aperture ring,and ofcourse had to be rechipped so it will work on digital cameras.

[mysticcowboy]

A lot of this thread is just silly. Comparing a K20D (or K7) with a d700 is absurd. They are built for different purposes and markets. And saying that the weight difference is trivial obviously comes from someone who hasn't carried a d700 and a 2.8 70-200 lens around for a few hours.

Auto focus isn't everything, says this graying guy who learned on fully manual cameras.

Folks, you don't need to be defensive for shooting a Pentax. I switched because I only wanted to pay for image stabilization once. So what that Pentax doesn't make a great action sport model body. The K7 and K20D are superb general purpose/wedding/landscape cameras. I don't shoot action sports. I don't need the extra weight and expense of a Nikon. General IQ on both systems is excellent.

[Biro]

Before I bought my K200D, I had a Nikon D200 for a short while. I quickly decided that not only was it too big and too heavy for my liking... but that I spent too much time thinking about operating it instead of taking photographs. Nikons are great cameras... but they're not for everybody.

[Pål Jensen]

Frankly, I don't think the OP's first post proves anything. Even among the Pentax community, people claim Pentax bodies can't track focus as well as other brands.

I haven't found any subjects the AF cannot track. Regardless of brand. I find the AF issues totally baffling and every example I've seen that supposedly prove the issues has been obvious user errors (particularly over at DPreview). Mastering good technique for AF is in fact very difficult.
As for the low light AF "issue"; my K-7 use 1s to focus in a dark room with one lightbulb; that is out of range of meter light levels. AF work but metering doesn't.

[Christopher M.W.T]

Whilst I disagree that Pentax AF is superior in any way, it doesn't mean that it isn't good.

I was able to get some great shots with my 50-135 using AF-C and shooting as 6mp JPEGS as these shots prove.

http://www.flickr.com/photos/cmt83/sets/72157624058394542/

So yes under good lighting conditions the K-7's AF-C is great, but I've never shot with a Nikon or Canon enough to really know how much better it can be.

I do know though from my uses that the Nikons and Canons AF feels more responsive.

[Christopher M.W.T]

And these were in RAW AF-C centre point focus.

[ManuH]

Christopher, these subjects are not difficult at all for any AF system. In all AF discussions I've read, people always talk about different tasks. It's more difficult to track a toddler coming right to you at it is to track a car running fast but sideways at 100m. Look at the AF scale on your lens, on a DA* 16-50mm, from 2m/7ft to infinity there's almost no movement. It means that anything above this distance shoud be rather easy to AF with this lens. On the DA* 50-135mm, it's 10m/30ft. If you try difficult things like filling the frame at distances below 10m, the AF-C will have a very hard time to react fast enough. But even the best cameras have their limits, I think Canon publish these limits for their top models like the 1D series that are built for ultrafast AF and tracking.

[EricT]

Whilst I disagree that Pentax AF is superior in any way, it doesn't mean that it isn't good.

I was able to get some great shots with my 50-135 using AF-C and shooting as 6mp JPEGS as these shots prove.

Nile River Soccer Match - a set on Flickr

So yes under good lighting conditions the K-7's AF-C is great, but I've never shot with a Nikon or Canon enough to really know how much better it can be.

I do know though from my uses that the Nikons and Canons AF feels more responsive.

No offense, but I think any camera has fast enough AF to track this guy:

[PentaxPoke]

And these were in RAW AF-C centre point focus.

Sorry Christopher but those are poor examples of AF. Are you familiar with the concept of "depth of field?" It looks like the DOF in all your shots is so deep, as to be hyperfocal. Notice the last shot for example; the guy with the red hat and glasses on left is in perfect focus, as are the trees at infinity. I'm sorry to say that a fixed-focus disposable camera would have done just as well with the focus.

To really test AF speed and accuracy, you need a shallow DOF. The AF speed also strongly depends on available light (this is where Pentax lags. In low light AF speed.) To test tracking, you need something moving toward, or away from you (as Manu said). To really give AF a workout, you need all of the above.

[Vaikis_]

what about this shots ?
K-7 in AF.C mode + slow DA 55-300, last two with FA* 77 /F1.8
I'am shooting car drift events without big problems, would be good if Pentax made better AF.C then K-7 has now.

[PentaxPoke]

what about this shots ?
K-7 in AF.C mode + slow DA 55-300, last two with FA* 77 /F1.8
I'am shooting car drift events without big problems, would be good if Pentax made better AF.C then K-7 has now.

Were those supposed to be examples of good AF or bad AF ?