[Barry Pearson]

Hi! I'm pretty excited about the K-1, but one area where Pentax has been only adequate vs the competition is in AF. They make very accurate AF, but it's not really fast and the point spread is weird sometimes. I'd say my 645z keeps up with the old 5D Mark II, with a few extra points for low light accuracy. But my DSLR is the D750, which aside from the D5 probably has the best AF of any FF DSLR. How does the K-1 measure up? I shoot weddings so I need a camera to track slow moving subjects reasonably well, and lock focus at a dark reception reasonably quickly, ESPECIALLY with fast lenses. As an example, I had to upgrade to the D750 from the Df because the Df simply could not focus the 58mm 1.4G wide open, it would consistently miss. The D750 nails it every time, it's remarkable. But I'd switch to the K-1 in an instant if it offered even 75% of that performance.

Yesterday I shot birds with my K-1. Pretty well its first real use. Bird "portraits". Birds in flight. Yachts (!)

I have posted several photos, (in a number of separate posts), and observations, in a thread I started at the following:
Barry Pearson's K-1

I won't pretend I can extrapolate from this limited experience to your situation. But I'm sure you will get some clues, and perhaps I can answer further questions you may have. One post in that thread is:
Some comments on AF on the K-1

[falconeye]

But I'd switch to the K-1 in an instant if it offered even 75% of that performance.

I am sure the K-1 AF is good enough for 99% of use cases.

However, being curious myself, I did a quick and dirty side by side test. AF-S from out of focus into focus in normal store light:
1. K-1 + Pentax 70-200/2.8 at 200mm
2. D500 + Nikkor 70-200/2.8 at 200mm

Where, of course the D500 AF is that of the $7000 D5. I.e., this is an apples to oranges test. OTOH, I did not test AF-C where the D500 should excel.

Both combinations focussed with a noticeable lag, with the lag being 2 or 3 times larger for the K-1. I cannot say if this is from the AF or the lens focus motor. Don't get me wrong: the lag for the K-1 was still ok for most shooting situations. But it is not in the same class as D5/D500.

BTW, in live view, the D500 AF was ridiculously slow

[HopelessTogger]

I am sure the K-1 AF is good enough for 99% of use cases.

However, being curious myself, I did a quick and dirty side by side test. AF-S from out of focus into focus in normal store light:
1. K-1 + Pentax 70-200/2.8 at 200mm
2. D500 + Nikkor 70-200/2.8 at 200mm

Where, of course the D500 AF is that of the $7000 D5. I.e., this is an apples to oranges test. OTOH, I did not test AF-C where the D500 should excel.

Both combinations focussed with a noticeable lag, with the lag being 2 or 3 times larger for the K-1. I cannot say if this is from the AF or the lens focus motor. Don't get me wrong: the lag for the K-1 was still ok for most shooting situations. But it is not in the same class as D5/D500.

BTW, in live view, the D500 AF was ridiculously slow

Might see improvements with firmware updates. So long as the K-1 isn't an A7r I'll be very happy.

[ColiNiloK]

I'm also interested to find out in the shooting scenarios as the OP mentioned about the K-1's AF performance in wedding and event photography as I sometimes need to shoot those as well. Looking at the specs of K-1 AF on paper and from what I have read it should't be bad at all, in fact I think it might be better than cameras like 6D, 5D2, A99, A900 which I have used in shooting events and weddings and I know people who use them for shooting events and weddings as well.

If the K-1 AF is as good as D610 both using similar lenses (on paper it seem to have the closest AF spec to the K-1) then it should be more than enough to shoot slow moving actions. Just one have to work with the AF point coverage, then again even cameras know to have good AF like D5, D810, D750 and 5D3's AF points coverage isn't exactly much much bigger than the K-1's.

[phoebus]

Just tried K-1 with DA*200/2.8 and HD-DA 1.4x AW. After a few misses I found that AF.C with the selected point and tracking (see photo of rear screen below) worked pretty well. A completely different experience from the K-5 where only the dogs tail would have been in focus... For any that doubted it, the TC definitely is not FF capable with the 200.

Hopefully that demonstrates the K-1 is reasonable at tracking targets subjects moving towards the camera.

Photo taken with 1/1600 ISO400 280mm f 6.3

[Venom3300]

^did you have the lens hood on? That lens hood is designed for crop sensors.

[phoebus]

^did you have the lens hood on? That lens hood is designed for crop sensors.

Yes I did have hood on - good point...

...but just checked without hood and still vignetting. No TC and hood on; no vignetting. So hood is fine for FF, TC isn't :-(

[Venom3300]

Yes I did have hood on - good point...

...but just checked without hood and still vignetting. No TC and hood on; no vignetting. So hood is fine for FF, TC isn't :-(

Well rats, i was trying to save it lol

[phoebus]

Well rats, i was trying to save it lol

Yes, I'm disappointed with that too. Also just checked on F50/1.4 and Sigma 24-70/2.8 - vignetting was neglible except at wide end of the 24-70. I suspect that the DA* 200 is particularly susceptible as the rear element is wide and some distance recessed from the back of the lens, so the light rays are not diverging at much of an angle. The front element of the TC is relatively small.

I don't have the DA* 300, so I can't comment on whether that suffers as badly, but I would expect it to.

Hopefully a FF TC will be in the Ricoh master plan before too long.

[tomO2013]

Received my K1 yesterday. My current system(s) are an A7rii/A7ii and previously I shot 5d mark iii, EM1 and XE1.
I've also rented D810 for landscape/medium format replacement in the past and also used with wedding.

My take...
The focus system for AFS is shot to shot similar in experience to the D810. Ignoring the sheer volume of PDAF points, as DSLR's will never go the route of hundreds of points as we find on mirrorless - actually it would probably be disadvantageous if it comes at the expense of smaller cross type that are less sensitive. When it comes to DSLR's the points are not limited by the pixel size of the sensor plane. As they are off sensor they can be larger and more sensitive if required - probably partly why as pixel pitches increase on mirrorless we are likely to see larger and larger number of focus points on plane. It also makes for great marketing/selling brochure material "X camera are 500 focus points!". What I'm most interested in is the sensitivity, accuracy and the AF algorithms in place to make sensible/intelligent AF decisions. I feel that the AF algorithms is where Canon/Nikon have the advantage today in tracking over everybody.

Digress.... the K1 has more cross type than the D610 and the focus spread is wider. Focus speed for me is equivalent to the D810 / 5d mark iii in point to point (using SEL point center). It's faster with the 24-70 in good light. It's also quite a bit more sensitive/accurate in lower light than I have experienced with the D810 or A7rii.

The bad news on AF is that I have found in lower light and pushing the af system e.g. -1.5 -> -2EV and wide area focus, ISO3200-6400, the K1 with 24-70 2.8 the camera hesitates as if it is looking for something to lock on to. It actually hesitates for longer than my A7ii (not a7rii) using firmware 3.1 or the D810.
This 'lag' for a few moment is a little frustrating but does eventually lock focus very accurately. The k3ii did not exhibit this behavior and was actually much quicker in this scenario leading me to think it's a bug or firmware update. With center single point af it's much better but can still exhibit the same behaviour to a lesser extent.

Where this is weird for me is that swapping over to live view and the K1's CDAF I've actually gotten a faster result. If Pentax can remove the need to reset the mirror after each shot when in live view, then in theory this is not far off mirrorless speed! I would expect the PDAF and OVF to be faster here!

This is the only negative that I can speak to right now regarding the AF and I have not noticed this with the 77mm limited so it's likely only a firmware bug that Pentax will fix.

Other than that the camera is super responsive and in most lighting situations (to echo falcon eye's point) it is plenty accurate for 99% of working/shooting scenarios.

[bobbotron]

I've always thought it was a bit odd with the K3 that the mirror did anything during live view shutter operation. I'm sure there's a good reason for it?

Received my K1 yesterday. My current system(s) are an A7rii/A7ii and previously I shot 5d mark iii, EM1 and XE1.
I've also rented D810 for landscape/medium format replacement in the past and also used with wedding.

My take...
The focus system for AFS is shot to shot similar in experience to the D810. Ignoring the sheer volume of PDAF points, as DSLR's will never go the route of hundreds of points as we find on mirrorless - actually it would probably be disadvantageous if it comes at the expense of smaller cross type that are less sensitive. When it comes to DSLR's the points are not limited by the pixel size of the sensor plane. As they are off sensor they can be larger and more sensitive if required - probably partly why as pixel pitches increase on mirrorless we are likely to see larger and larger number of focus points on plane. It also makes for great marketing/selling brochure material "X camera are 500 focus points!". What I'm most interested in is the sensitivity, accuracy and the AF algorithms in place to make sensible/intelligent AF decisions. I feel that the AF algorithms is where Canon/Nikon have the advantage today in tracking over everybody.

Digress.... the K1 has more cross type than the D610 and the focus spread is wider. Focus speed for me is equivalent to the D810 / 5d mark iii in point to point (using SEL point center). It's faster with the 24-70 in good light. It's also quite a bit more sensitive/accurate in lower light than I have experienced with the D810 or A7rii.

The bad news on AF is that I have found in lower light and pushing the af system e.g. -1.5 -> -2EV and wide area focus, ISO3200-6400, the K1 with 24-70 2.8 the camera hesitates as if it is looking for something to lock on to. It actually hesitates for longer than my A7ii (not a7rii) using firmware 3.1 or the D810.
This 'lag' for a few moment is a little frustrating but does eventually lock focus very accurately. The k3ii did not exhibit this behavior and was actually much quicker in this scenario leading me to think it's a bug or firmware update. With center single point af it's much better but can still exhibit the same behaviour to a lesser extent.

Where this is weird for me is that swapping over to live view and the K1's CDAF I've actually gotten a faster result. If Pentax can remove the need to reset the mirror after each shot when in live view, then in theory this is not far off mirrorless speed! I would expect the PDAF and OVF to be faster here!

This is the only negative that I can speak to right now regarding the AF and I have not noticed this with the 77mm limited so it's likely only a firmware bug that Pentax will fix.

Other than that the camera is super responsive and in most lighting situations (to echo falcon eye's point) it is plenty accurate for 99% of working/shooting scenarios.

[clackers]

.

Digress.... the K1 has more cross type than the D610 and the focus spread is wider. Focus speed for me is equivalent to the D810 / 5d mark iii in point to point (using SEL point center). It's faster with the 24-70 in good light. It's also quite a bit more sensitive/accurate in lower light than I have experienced with the D810 or A7rii..

This probably needs to be a sticky. :-P

[noelpolar]

For what it's worth, K-1 BIF is much much better then with my K3 (both with DFA 150-450)

[walro]

My take on the af speed after several hours with the camera. In certain situations it is not that much faster, but others it beats all other Pentax cameras I've ever used. In particular, using a.f.c, I was able to get almost all shots of children swinging and going down slides in focus. Whereas before I would be happy to get one or two in focus out of a series of 5 or 6.

I shot a lot with the 200 without the hood, and had no vignetting whatsoever. I haven't tried it with the hood. That lens is a dream on the K1.

[tomO2013]

Ok further update....

I think I've narrowed this weird low light behaviour down to PDAF focus settings in AFS. I'm convinced a firmware update will quietly update this. And I have confirmed behaviour on two shop bodies.
If you enable AF-S but have focus priority (even with center point) enabled the camera will focus to the point, hesitate frustratingly, then light up the focus point LED and release the shutter. This hesitation is crazy because of the aforementioned post with CDAF being faster in live view than OVF and PDAF.

Switching the PDAF and AFS to release priority and it's an entirely different ball game and makes a huge difference to focus performance and camera operation for event work. A half depress has the camera racing to focus and a full press has it light the focus LED and grab the shot. Even test shots going from infinity to close with a full depress of the shutter in this mode have yielded in focus and accurate shots. It's blazing quick with 24-70, 15-30, the 3 amigos and even 70-200.

Tonight I've done a side by side with a buddy with a D750 (set to release priority as well as well as focus priority) in single shot AFS.
In the ISO6400 and low light (possibly -2EV) the K1 was both noticeably faster and more accurate when both had their focus point set to center af point. Pentax with the DFA* 70-200 and Nikkor 70-200 F2.8 on the D750. Both wide open at F2.8.

For now I've my camera set to release priority and I'm suddenly really enjoy this camera a lot - it's focusing is pro-grade DSLR quick