lens or body faulty?
Posted 15/05/2008 - 13:20
Link
My K10D back-focussed exactly the same way
You use the 45 degree one and select centre focus. That way (as in your case) it focusses on the big (unmissable) black bar in the middle. With your example you read off the numbers and check what's in focus - around 30 to 40mm back is what I see.
If all your lenses behave this way, then you'll need to adjust the K10D focus (you need a v1.1 firmware... see other threads here).
Took me about half hour to tweak mine and the K10D has been a different camera ever since
Matt
You use the 45 degree one and select centre focus. That way (as in your case) it focusses on the big (unmissable) black bar in the middle. With your example you read off the numbers and check what's in focus - around 30 to 40mm back is what I see.
If all your lenses behave this way, then you'll need to adjust the K10D focus (you need a v1.1 firmware... see other threads here).
Took me about half hour to tweak mine and the K10D has been a different camera ever since
Matt
http://www.mattmatic.co.uk
(For gallery, tips and links)
(For gallery, tips and links)
Posted 15/05/2008 - 13:34
Link
thanks Matt, i took it to the shop and they managed to get a good image of the chart (somehow) so they declared the camera good and pronounced my technique (camera shake) as the culprit!...500th second, I think not.
I didn't want to fiddle with the debug menu until I had given them a chance to look at it.
After they had it, my card had three unusable (out of focus) images of outside their shop then a suspicious gap of 28 images then two reasonable shots of the chart!
my other AF lenses exhibit a degree of back focus but not as much.
Is it possible to set it up for different lenses? could the lens itself cause this?
any thoughts on my wonky picture? is this normal? (happens everytime)
I didn't want to fiddle with the debug menu until I had given them a chance to look at it.
After they had it, my card had three unusable (out of focus) images of outside their shop then a suspicious gap of 28 images then two reasonable shots of the chart!
my other AF lenses exhibit a degree of back focus but not as much.
Is it possible to set it up for different lenses? could the lens itself cause this?
any thoughts on my wonky picture? is this normal? (happens everytime)
Posted 15/05/2008 - 20:29
Link
Yeah that did occur to me, but this is after the shop visit. I asked them to check the body for back focus, but they didn't appear to test it with any other lens.
They were a joke, they kept my camera and 16-45 and told me they would get back to me. took a bunch of pictures outside - all out of focus, then 25 shots went missing then they went inside and tried the printed sheet test. like I said they supposedly managed a couple of good shots but I wasn't there and I haven't managed anything in focus since getting it back. Their comment in front of other customers that my test shots in the park were blurred by camera shake really pissed me off, as all the shots were a minimum of 250th second and all showed areas in focus behind the subject.
How easy is it to adjust the camera and how likely is it to be just the lens as my other lenses seem to be less affected. (tolerable mostly).
BTW I took the lens in to them a couple of weeks ago and it came back 'fixed' but it seems worse if anything.
They were a joke, they kept my camera and 16-45 and told me they would get back to me. took a bunch of pictures outside - all out of focus, then 25 shots went missing then they went inside and tried the printed sheet test. like I said they supposedly managed a couple of good shots but I wasn't there and I haven't managed anything in focus since getting it back. Their comment in front of other customers that my test shots in the park were blurred by camera shake really pissed me off, as all the shots were a minimum of 250th second and all showed areas in focus behind the subject.
How easy is it to adjust the camera and how likely is it to be just the lens as my other lenses seem to be less affected. (tolerable mostly).
BTW I took the lens in to them a couple of weeks ago and it came back 'fixed' but it seems worse if anything.
Posted 16/05/2008 - 06:12
Link
The appearance of back focus will seem different based on aperture. So, if you test with a 50mm f/1.4 wide open at f/1.4 you'll really, really see any back focus issues (that's how I sorted mine - which also showed that shooting with the 12-24 was also backfocussing and is now fixed).
With the K10D you can't adjust on a per-lens basis, but given the above point it's unlikely that you'll need to
If you were closer, I'd say pop by and I'd sort it out for you (takes 30min max)... if you're ever in East Sussex, drop me a PM
Matt
With the K10D you can't adjust on a per-lens basis, but given the above point it's unlikely that you'll need to
Quote:
How easy is it to adjust the camera and how likely is it to be just the lens as my other lenses seem to be less affected. (tolerable mostly).
It's not difficult really. Just a bit fiddly (mostly because no one seems to know how to adjust with the new 1.3 firmware So you have to get v1.1 back in the camera... which is a little tricky, alas).How easy is it to adjust the camera and how likely is it to be just the lens as my other lenses seem to be less affected. (tolerable mostly).
If you were closer, I'd say pop by and I'd sort it out for you (takes 30min max)... if you're ever in East Sussex, drop me a PM
Matt
http://www.mattmatic.co.uk
(For gallery, tips and links)
(For gallery, tips and links)
Posted 16/05/2008 - 06:30
Link
EDIT: Did some digging and found this link on DPReview.
Originally from RiceHigh's blog.
You may find this quote useful:
Apparently Pentax have their own software that allows focus adjustment of up to four lenses, but this little "hack" just enables and disables the Debug mode via USB (perhaps they completely removed the awkward button option!)
I'll keep you posted
Matt
Originally from RiceHigh's blog.
You may find this quote useful:
Quote:
I've done a quick test, and YES, it works!
For the English speaking world, the button captions will have question marks if you don't have the right language packs installed, but...
Connect to the camera with the upper left button.
Select the bottom radio button and click the button below it.
Disconnect with the button below the upper left button.
You're in Debug mode!
Reverse the above to exit Debug.
The far right bottom button is Exit.
Very nice!
As usual this info is provided AT YOUR OWN RISK, etc, etc, etc (For Vista users, read this thread)I've done a quick test, and YES, it works!
For the English speaking world, the button captions will have question marks if you don't have the right language packs installed, but...
Connect to the camera with the upper left button.
Select the bottom radio button and click the button below it.
Disconnect with the button below the upper left button.
You're in Debug mode!
Reverse the above to exit Debug.
The far right bottom button is Exit.
Very nice!
Apparently Pentax have their own software that allows focus adjustment of up to four lenses, but this little "hack" just enables and disables the Debug mode via USB (perhaps they completely removed the awkward button option!)
I'll keep you posted
Matt
http://www.mattmatic.co.uk
(For gallery, tips and links)
(For gallery, tips and links)
Posted 16/05/2008 - 11:50
Link
Thaks Matt, I'll check this out and get back to you.
Posted 16/05/2008 - 13:02
Link
The general setup is this:
1. Get the fastest lens you have (an f/1.7 or f/1.4 is a good choice)
2. Use the focus chart
3. Set the camera on a tripod, pointing at 45 degrees to the chart
4. Get the K10D into debug mode
5. Using the AF focus adjust menu, make a note of the current adjustment (measured in "nm" I believe).
6. Use the adjustment to shift the focus - try a smallish value (say 40) first.
7. Take a test shot and zoom into the playback to see what numbers are in focus on the RHS of the shot.
8. Adjust the focus and repeat until happy
9. Once happy, turn off the debug mode
You can try with other lenses, and there may be a small amount of variability between lenses, but you should be able to get it far better than you are showing now!
Matt
1. Get the fastest lens you have (an f/1.7 or f/1.4 is a good choice)
2. Use the focus chart
3. Set the camera on a tripod, pointing at 45 degrees to the chart
4. Get the K10D into debug mode
5. Using the AF focus adjust menu, make a note of the current adjustment (measured in "nm" I believe).
6. Use the adjustment to shift the focus - try a smallish value (say 40) first.
7. Take a test shot and zoom into the playback to see what numbers are in focus on the RHS of the shot.
8. Adjust the focus and repeat until happy
9. Once happy, turn off the debug mode
You can try with other lenses, and there may be a small amount of variability between lenses, but you should be able to get it far better than you are showing now!
Matt
http://www.mattmatic.co.uk
(For gallery, tips and links)
(For gallery, tips and links)
Posted 16/05/2008 - 15:04
Link
I have what might seem like a stupid question, but I'm not sure how/if this affects the results...
On my camera, the center auto focus point (actually all of them) appear to be not exactly centered. If I take a blank piece of paper with a black dot on it and move it around, I can find the edges and shape of each auto focus point. With that I've found that my focus points were down about 15% of the frame and a tad to the left.
1. Does that seem a valid test?
2. If the focus point is not centered, will the focus chart be a valid test for focus adjusting? i.e. is the focus on the far edge of the center bar or the near edge?
Does it matter? If the camera focuses properly does it matter much exactly where the focus point is?
My guess, is if the camera is focusing properly, but the focus point is not where I think it should be, I as the viewer will think it is off. Adjusting the focus depth would only make things worse.
I'm thinking of making the adjustments as well, but not sure if I'm really going to do much better if the camera is a bit out of whack to begin with.
Not sure it is worth another ten weeks to send for repair.
On my camera, the center auto focus point (actually all of them) appear to be not exactly centered. If I take a blank piece of paper with a black dot on it and move it around, I can find the edges and shape of each auto focus point. With that I've found that my focus points were down about 15% of the frame and a tad to the left.
1. Does that seem a valid test?
2. If the focus point is not centered, will the focus chart be a valid test for focus adjusting? i.e. is the focus on the far edge of the center bar or the near edge?
Does it matter? If the camera focuses properly does it matter much exactly where the focus point is?
My guess, is if the camera is focusing properly, but the focus point is not where I think it should be, I as the viewer will think it is off. Adjusting the focus depth would only make things worse.
I'm thinking of making the adjustments as well, but not sure if I'm really going to do much better if the camera is a bit out of whack to begin with.
Not sure it is worth another ten weeks to send for repair.
Posted 16/05/2008 - 15:42
Link
I was wondering the same thing, that's why I posted an image with centre of image and focus indicator point marked on them.
But the perpendicular image I used for levels was way out of focus also, and that wouldn't be affected by the position of the centre point, so my camera is definately not focussing properly.
I took a picture of a camera and the spirit level while I was there and they both showed focus was behind.
I think the chart is definately flawed if not used with caution.
But the perpendicular image I used for levels was way out of focus also, and that wouldn't be affected by the position of the centre point, so my camera is definately not focussing properly.
I took a picture of a camera and the spirit level while I was there and they both showed focus was behind.
I think the chart is definately flawed if not used with caution.
Posted 16/05/2008 - 21:01
Link
Huh?
That's why the centre black bar is so large - to filter out any focus point offset. The focus test chart at 45 degrees has such a huge and unmissable central target that the camera just has to get it.
Ok, so the focus point may be a fraction out (looks less than 1mm for that view), but looking at the 45 degree shot the K10D was a full 40mm back-focussed
That's definitely not to do with where the focus point sits in the viewfinder... that's back-focus error.
(And in fact, it was about the same amount as mine before it was adjusted. As I said, even the 12-24 at 12mm at f/8 you could see it had back-focussed even on a landscape! Having adjusted for a f/1.4 it's great on 12mm )
Matt
That's why the centre black bar is so large - to filter out any focus point offset. The focus test chart at 45 degrees has such a huge and unmissable central target that the camera just has to get it.
Ok, so the focus point may be a fraction out (looks less than 1mm for that view), but looking at the 45 degree shot the K10D was a full 40mm back-focussed
That's definitely not to do with where the focus point sits in the viewfinder... that's back-focus error.
(And in fact, it was about the same amount as mine before it was adjusted. As I said, even the 12-24 at 12mm at f/8 you could see it had back-focussed even on a landscape! Having adjusted for a f/1.4 it's great on 12mm )
Matt
http://www.mattmatic.co.uk
(For gallery, tips and links)
(For gallery, tips and links)
Posted 16/05/2008 - 21:04
Link
I'm firing off the program and delving into the debug menu right now... wish me luck!
Posted 16/05/2008 - 22:16
Link
OH MY GOD!!!
i HAVE A NEW LENS...OOPS CAps lock...
bloody 'ell, why is there so much mystery surrounding this? that has totally transformed my lens ( I say lens, cos I may have buggered up all the others )
I went up in 30s until I reached 120 (at which point I had to move the tripod back to get focus) and now this lens is embarrassingly sharp! (sorry Claire... there's always Gaussian blur )
I now have about 10mm in front and 40mm behind)
OK, now to try the other lenses..gulp.
Is there any reason to take it out of debug mode?
i HAVE A NEW LENS...OOPS CAps lock...
bloody 'ell, why is there so much mystery surrounding this? that has totally transformed my lens ( I say lens, cos I may have buggered up all the others )
I went up in 30s until I reached 120 (at which point I had to move the tripod back to get focus) and now this lens is embarrassingly sharp! (sorry Claire... there's always Gaussian blur )
I now have about 10mm in front and 40mm behind)
OK, now to try the other lenses..gulp.
Is there any reason to take it out of debug mode?
Posted 17/05/2008 - 01:59
Link
OMG!! I have been kicking myself for having poor focus skills but after running through this adjustment, I am really looking forward to using my camera!
The test chart witha 45 dgree angle doesnt' show a nice 30/60 split in front/back DOF, but for real life captures -- like people say, "its a whole new camera"!
Now I have to get used to focusing correctly. After a year and a half I've come to adjust to what I've been seeing. (thought it was my eyes and shaky hands. Ugh!!! :roll
Check out the before and after (ignore the wonky color balance)
I had to set mine to -170.

The test chart witha 45 dgree angle doesnt' show a nice 30/60 split in front/back DOF, but for real life captures -- like people say, "its a whole new camera"!
Now I have to get used to focusing correctly. After a year and a half I've come to adjust to what I've been seeing. (thought it was my eyes and shaky hands. Ugh!!! :roll
Check out the before and after (ignore the wonky color balance)
I had to set mine to -170.

Posted 17/05/2008 - 02:35
Link
Not sure if it is clear in this image...
And maybe my testing is flawed, but whether you focus with the line at the top of the AF area or at the bottom of the AF area does make a difference.
Of course this movement adjusts distance, relative size, etc... but all of that is part of making the image in the first place and in my opinion still applies to the issue and not sure this would invalidate the test.
If your AF sensor areas are not in the right place, the final image will be affected with regards to DOF in front or in back of the exact point you are hoping to focus on.
As long as I know where my AF point is, I can be consistent. It is simply important in some limited cases to know where that is. Lets say for macro for instance.
The top image I focused with the black line at the top of the AF area (for my camera is nearly dead center). For the bottom image I focused with the black bar at the bottom of the AF area (a bit below the expected AF area).
Still, I understand what you were saying... it certainly isn't any 40mm difference.

And maybe my testing is flawed, but whether you focus with the line at the top of the AF area or at the bottom of the AF area does make a difference.
Of course this movement adjusts distance, relative size, etc... but all of that is part of making the image in the first place and in my opinion still applies to the issue and not sure this would invalidate the test.
If your AF sensor areas are not in the right place, the final image will be affected with regards to DOF in front or in back of the exact point you are hoping to focus on.
As long as I know where my AF point is, I can be consistent. It is simply important in some limited cases to know where that is. Lets say for macro for instance.
The top image I focused with the black line at the top of the AF area (for my camera is nearly dead center). For the bottom image I focused with the black bar at the bottom of the AF area (a bit below the expected AF area).
Still, I understand what you were saying... it certainly isn't any 40mm difference.

Add Comment
To leave a comment - Log in to Pentax User or create a new account.


1753 posts
18 years
Plymouth,
UK
This shot was taken from a tripod with everything perfectly square in the view finder and auto-focused.
several shots all display wonky like this. and all shots display apparent back focus.
this one shows the test chart at 45 degrees. red focus indicator went off where the green box has been drawn. but the red lines show the exact centre of the image. I waited for the green focus light (in both cases).
I have a whole load more, if these are inconclusive.
Plymouth Photographer