Isolating Focus problems with the K5

[Paratom]

Here is the problem:

I am on the second K5, I own 15,21,35Macro,70 Ltd. lenses and the 50-135 (and the 17-55 which I dont use because IMO its not worth it)

I have finetuned AF a lot back and forth with those lenses because I have had problems with AF. Now I am back to zero or max +/- 2 adjustment.

Short distance focus seems ok, not super accurate like my Nikon but not so bad.
The main problem starts when we have distances around 2meters and longer.

Question 1:
To find out if I am alone could maybe anybody here shoot a leaf lying on the grass from 1,2,3,5meters distance wide open with a focal length 35mm or longer?

Theory 1:
Now I also have one theory: the focus thread on the Limited DA primes is very very short. The 2m "position" of the focus thread is very close to the 3m to the 5m to infinity. Turning the focus ring 1 degree will lead to a much bigger change in distance compared to other lenses from other brands I own.
So maybe this could cause a certain innaccurancy in focusing???
So maybe a lens with a longer focus thread could be focused more accurate?

Question 2:
I have the AF set on the AF button
Does anybody here know if it makes a difference if I set it to AF-S or to AF-C in this position. AF-S locks and stay in the same position, AF-C sometimes is focusing back and forth a little (so which position would be the right one???) (however the Canon 7D for example have the same tendancy, the Nikon D700 seems to "lock" better even in AFC position).
So any idea if AF-S should be more reliable for static subjects?

Medium distance is where I really have the problem.

Thanks for any ideas.

 

[jonoslack]

Hi Tom
Busy right now, but I'll give it a try.
Just as a matter of interest - if you set focusing back to the half press on the shutter - does that make any difference? (it's the one thing I can see that you do really differently from me).

Incidentally - I've switched off all the focus adjustment on mine too.

Is it also true of your 50-135?

all the best

 

[Paratom]

Hi Tom
Busy right now, but I'll give it a try.
Just as a matter of interest - if you set focusing back to the half press on the shutter - does that make any difference? (it's the one thing I can see that you do really differently from me).

Incidentally - I've switched off all the focus adjustment on mine too.

Is it also true of your 50-135?

all the best

I havent tried to put it on the shutter lately. Will give it a try.

Yes, also my 50-135 I get inconsistent focus at medium-long distance. (Here my short focus thread theory wouldnt be valid)

 

[DavidL]

I've just done this quickly with 40mm f2.8 lens at f2.8
It's a flower on a wall I found that easier than a leaf on the grass as, as I got over 3m the leaf was too small. I know should have got a larger leaf
Does this help?
removed images and posted a new set.

David

 

[Paratom]

Thank you very much David.
The second image looks like 1-2 stones closer to the photographer might be sharper than the flower.
Also the 100% crop doesnt really look sharp to me (in the second).
First shot looks pretty fine to me.

What do you think?

 

[JMaher]

I am interested as well. I agree the first shot looks great but the 2nd & 3rd crops are both not as sharp as I would like. As I understand it the focus point is fairly large and locks onto the highest contrast. If that's true it would be hard to focus on something as small as the flower at any distance and have it be in focus all the time.

I am not having any problem with focus but I am not concentrating on a small object at distance. More likely a person which is obviously a much larger target.

Tom, you mentioned the D700. I have shot with this and I suspect there may be no better camera to lock in focus quickly and consistently on the planet. However there are many things I like better about the K5.

Jim

 

[limbonaut]

I had the same problems and ultimately decided to sell my K-5 and go back to Sony, but even though I have zero focus problems with the A580 and the image quality is similar I really miss the ergonomics and handling of the K-5. I tested many lenses and under many circumstances but always had inconsistent focus on all 3 of my K-5 bodies. The most frustrating thing was seeing how in focus shots using live view were.

I'd love to hear that more recently manufactured bodies do not have focus issues, but I am beginning to think the K-5's system is either not performing as designed or just not a very good design.

Here is the problem:

I am on the second K5, I own 15,21,35Macro,70 Ltd. lenses and the 50-135 (and the 17-55 which I dont use because IMO its not worth it)

I have finetuned AF a lot back and forth with those lenses because I have had problems with AF. Now I am back to zero or max +/- 2 adjustment.

Short distance focus seems ok, not super accurate like my Nikon but not so bad.
The main problem starts when we have distances around 2meters and longer.

Question 1:
To find out if I am alone could maybe anybody here shoot a leaf lying on the grass from 1,2,3,5meters distance wide open with a focal length 35mm or longer?

Theory 1:
Now I also have one theory: the focus thread on the Limited DA primes is very very short. The 2m "position" of the focus thread is very close to the 3m to the 5m to infinity. Turning the focus ring 1 degree will lead to a much bigger change in distance compared to other lenses from other brands I own.
So maybe this could cause a certain innaccurancy in focusing???
So maybe a lens with a longer focus thread could be focused more accurate?

Question 2:
I have the AF set on the AF button
Does anybody here know if it makes a difference if I set it to AF-S or to AF-C in this position. AF-S locks and stay in the same position, AF-C sometimes is focusing back and forth a little (so which position would be the right one???) (however the Canon 7D for example have the same tendancy, the Nikon D700 seems to "lock" better even in AFC position).
So any idea if AF-S should be more reliable for static subjects?

Medium distance is where I really have the problem.

Thanks for any ideas.

 

[DavidL]

I rushed the previous shots, these not so much. This sort of testing can create all sorts of problems some real, some imagined and some caused by carelessness in my case. Anyway I did some more.
The problem I experienced was as you get farther away from what your focusing on there is less definition to that part of the subject anyway. The flower head was a bad choice.
The focusing on the K5 isn't in the same league as Nikons, in my opinion, so I'm not sure you'll ever be satisfied. My Nikon's have gone, so I have to be otherwise:banghead:

 

[Sapphie]

There's a lot more to this than meets the eyeball. How are we judging sharpness/focus? I too have had 'issues' with my K5, most, probably, my own fault/delusion/imagination. My GH2 (now departed) and my X100 seemed initially 'sharper', on-screen. The few prints (12x8) I have made from my K5 reminded me of prints from a by-gone film era (a good thing) ...

Lee

 

[Paratom]

I rushed the previous shots, these not so much. This sort of testing can create all sorts of problems some real, some imagined and some caused by carelessness in my case. Anyway I did some more.
The problem I experienced was as you get farther away from what your focusing on there is less definition to that part of the subject anyway. The flower head was a bad choice.
The focusing on the K5 isn't in the same league as Nikons, in my opinion, so I'm not sure you'll ever be satisfied. My Nikon's have gone, so I have to be otherwise:banghead:

those look totally fine to me.

 

[Paratom]

There's a lot more to this than meets the eyeball. How are we judging sharpness/focus? I too have had 'issues' with my K5, most, probably, my own fault/delusion/imagination. My GH2 (now departed) and my X100 seemed initially 'sharper', on-screen. The few prints (12x8) I have made from my K5 reminded me of prints from a by-gone film era (a good thing) ...

Lee

I judge focus by plane of sharpness. If I focus on eyes of a face I want them sharp.

 

[raist3d]

I can corroborate pretty much all the af issues mentioned. Also how accurate the cdaf usually is. In general what I find is that it boils down to how big the focus areas are. Anything where the object is thinner you risk the camera focusing on something else as the background or foreground has more detail.

Given how finicky af in general seems to be with cameras perhaps except nikon and I still hear from a very experienced nikon photographer you can still miss I am learning to do manual focus.

This is an area I expect microfour thirds to be doing better already for the most part.

- raist

 

[Paratom]

I can corroborate pretty much all the af issues mentioned. Also how accurate the cdaf usually is. In general what I find is that it boils down to how big the focus areas are. Anything where the object is thinner you risk the camera focusing on something else as the background or foreground has more detail.

Given how finicky af in general seems to be with cameras perhaps except nikon and I still hear from a very experienced nikon photographer you can still miss I am learning to do manual focus.

This is an area I expect microfour thirds to be doing better already for the most part.

- raist

Raist,
now if you are going to manual focus ( and I do this sometimes when shootning the S2) a large viewfinder is quite an advantage.
So IMO for manual focus I would probably want to go FF, for example the Sony A900 or Nikon or Canon and if I went manual focus I would probably also use Zeiss lenses.

The AF problems I have is not only for supersmall subjects, but also for faces for example at 3-5m distance.
The Nikon nails it nearly all the time.I cant believe its that much better and still searching if I do something wrong.

 

[woodyspedden]

Only problem with this Tom is that you end up with a system that is as large or larger than your S2 and quite a bit heavier.

I was shooting the D3 and D3X prior to getting my S2 and once the S2 and 70mm arrived it was clear that I would put up with the S2 because of IQ but I needed a smaller lighter system as well for everyday use. I bought the K-5 and have not looked back although I too have experienced the occasional focus issue, particularly with the 31FA.

Woody

 

[Paratom]

Only problem with this Tom is that you end up with a system that is as large or larger than your S2 and quite a bit heavier.

I was shooting the D3 and D3X prior to getting my S2 and once the S2 and 70mm arrived it was clear that I would put up with the S2 because of IQ but I needed a smaller lighter system as well for everyday use. I bought the K-5 and have not looked back although I too have experienced the occasional focus issue, particularly with the 31FA.

Woody

Hi Woody,
for me personally I dont want to use my DSLR in manual focus mode.
The DSLR I need for a) when I want faster focus or b) a smaller camera than my S2.
Everything bigger than a D700 doesnt make much sense for me.
The K5 is even (much) better regarding size.

 

[jonoslack]

I think I must be a real oddity.:loco:

I'm the guy who sent back 5 copies of the Pentax 16-50
3 copies of the Nikon 17-55
. . . . and all my Sigma lenses have been fine.

I had AF problems with my Nikon D700 with areas of high contrast behind the subject.

But focusing with the K5 just seems to be okay . . . . . but perhaps it's the lenses I'm using?:

18-135 (75%)
100 WR (10%)
35 macro (10%)
60-250 (4%)
ZK 50 f1.4 (1%)

Or perhaps I'm just a very special and exclusive advocate of Sod's Law!
Sure, I do get the odd out of focus shot - but it's scarce, and I can do that with any camera.

 

[JMaher]

Not the only oddity. The vast majority of my shots are in focus. I can't say as much for my artistic ability but they are in focus. No problem with people. I do use the focus point selector and single point focus most of the time. I also use the half press front button instead of the rear button.

I have a higher in focus percentage with the K5 than with a Canon 5D2. My memory of a D700 says even more would be in focus under demanding use. I would not expect to see a difference under normal use. All my lenses (after trying lots of options) are set at "0" for focus adjustment

I just checked LR for my use:
18-135 (61%)
50 1.7 (24%)
70 (15%)

Jim

 

[tsjanik]

Here is the problem:

I am on the second K5, I own 15,21,35Macro,70 Ltd. lenses and the 50-135 (and the 17-55 which I dont use because IMO its not worth it)

I have finetuned AF a lot back and forth with those lenses because I have had problems with AF. Now I am back to zero or max +/- 2 adjustment.

Short distance focus seems ok, not super accurate like my Nikon but not so bad.
The main problem starts when we have distances around 2meters and longer.

Question 1:
To find out if I am alone could maybe anybody here shoot a leaf lying on the grass from 1,2,3,5meters distance wide open with a focal length 35mm or longer?

Theory 1:
Now I also have one theory: the focus thread on the Limited DA primes is very very short. The 2m "position" of the focus thread is very close to the 3m to the 5m to infinity. Turning the focus ring 1 degree will lead to a much bigger change in distance compared to other lenses from other brands I own.
So maybe this could cause a certain innaccurancy in focusing???
So maybe a lens with a longer focus thread could be focused more accurate?

Question 2:
I have the AF set on the AF button
Does anybody here know if it makes a difference if I set it to AF-S or to AF-C in this position. AF-S locks and stay in the same position, AF-C sometimes is focusing back and forth a little (so which position would be the right one???) (however the Canon 7D for example have the same tendancy, the Nikon D700 seems to "lock" better even in AFC position).
So any idea if AF-S should be more reliable for static subjects?

Medium distance is where I really have the problem.

Thanks for any ideas.

I just recieved my K5 and having noted this thread I was somewhat concerned about the AF. In my limited experience, in matrix mode the camera can become confused. I changed to central spot focus and enabled AF on the shutter release (half press); using AF-S, I focus where desired and then adjust position for the compostion while maintaining the half press of the shutter, which locks the focus. Seems to work fine so far. So you might try that approach with static subjects. I apologize if this has been suggested before, I haven't followed this problem too closely since I had no intention of getting a K5 until I started following the Fun with the K-5 thread

Tom

 

[raist3d]

Raist,
now if you are going to manual focus ( and I do this sometimes when shootning the S2) a large viewfinder is quite an advantage.
So IMO for manual focus I would probably want to go FF, for example the Sony A900 or Nikon or Canon and if I went manual focus I would probably also use Zeiss lenses.

The A900 has one of the very best view finders I have ever seen and allows full manual focus even with F2.8 and close. I tried it out for myself. But the A900 as much as I also love its ergonomics, it's still a big heavy camera. I like small and light. If I can learn to do it with the K-5 even if it's harder, I rather do it on the K-5.

The AF problems I have is not only for supersmall subjects, but also for faces for example at 3-5m distance.
The Nikon nails it nearly all the time.I cant believe its that much better and still searching if I do something wrong.

Well it depends. Correct me if I am wrong but I think you mentioned focus on the eye of a face. That counts as a small object. Basically anytime there's anything of an area bigger than the sensor around your object, and has detail, there's a good chance the camera will go there. If I wanted to AF and I can take my time, I use contrast AF in that situation.

Of of course, MF + live view

Nikon is king of the hill for AF in the market, so yeah, NIkon's the best in that area.

- Raist

 

[raist3d]

I think I must be a real oddity.:loco:

I'm the guy who sent back 5 copies of the Pentax 16-50
3 copies of the Nikon 17-55
. . . . and all my Sigma lenses have been fine.

I had AF problems with my Nikon D700 with areas of high contrast behind the subject.

But focusing with the K5 just seems to be okay . . . . . but perhaps it's the lenses I'm using?:

18-135 (75%)
100 WR (10%)
35 macro (10%)
60-250 (4%)
ZK 50 f1.4 (1%)

Or perhaps I'm just a very special and exclusive advocate of Sod's Law!
Sure, I do get the odd out of focus shot - but it's scarce, and I can do that with any camera.

I would say check the distances at which you are focusing and how the objects show up %tage wise of the view finder. Also in what light.

- Raist