Forgot Password
Pentax Camera Forums Home
 

Closed Thread
Show Printable Version Search this Thread
02-12-2011, 01:10 AM   #526
Veteran Member
creampuff's Avatar

Join Date: May 2007
Location: Singapore
Photos: Gallery
Posts: 3,953
QuoteOriginally posted by Smeggypants Quote
That's right. All DSLR cameras exhibit back and front focussing. That's why the good ones have parameters to calibrated each lens.

The K-5 front focussing in low light is a seperate issue though
In-body calibration isn't always the solution if a particular lens requires focus calibration. Some lenses can also exhibit a slight focus shift at close focusing distances.
Unless both camera and lens is properly calibrated, it is not always so clear cut that the camera is automatically at fault.

For example, I personally know of users of Sigma 30mm and 50mm lenses who had focus issues on their Canon and Nikon cameras. It can't be that the cameras had issues but that the lenses had to be adjusted by Sigma before they focused correctly. In Pentax's own case, the K10D had problems focusing with the DA 21mm Limited. I know of a good friend who tested more than a dozen copies of the DA 21mm on separate K10Ds and they all had focus errors. When tested on the K20D, the said lenses focused accurately.


Last edited by creampuff; 02-12-2011 at 01:26 AM.
02-12-2011, 02:30 AM   #527
New Member




Join Date: Jan 2011
Posts: 5
Hello, does anybody has the same af test cart?
Thank You
Attached Images
 
02-12-2011, 03:37 AM   #528
Veteran Member
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,871
QuoteOriginally posted by Christine Tham Quote
PentaxVersion 70.0.0
Pentax Model 0x12e76
DSP/CPU Firmware Version 1.01.00.05
Manufacture Date 2010-12-09
Internal Serial Number 30282

I can definitely confirm that if the AF assist light comes on, the end result is a focused shot. I have two shots (one with exposure 1.6sec, the other 2.0sec) that are perfectly focused.

Falk are you able to post your K-5's Model and Version numbers to make sure the camera build revisions? I don't know if these numbers will change if they make production changes, though.
Pentax Version: 70.0.0
Pentax Model: 0x12e76
DSP/CPU Firmware Version: 1.01.00.05
Manufacture Date 2010-10-11
Internal Serial Number 11030

So, except being two months younger, your camera looks exactly the same... Strange indeed...
02-12-2011, 03:44 AM   #529
Veteran Member
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,871
QuoteOriginally posted by win-win Quote
Hello, does anybody has the same af test cart?
Thank You
That kind of test chart is not useful at all.

You cannot know which portion of a subject is picked up by the AF sensor (which spawns a large region actually).

Either use a flat target perpendicular to the optical axis and measure blur (best and what I always do), or at least use a target with NO contrast in the middle third of the image (except for a line or edge which again must be perpendicular to the optical axis).

I've seen flawed focus tests several times and therefore, I thought to give advice on the topic. When using charts like above, you'll always obtain front focus even with a perfect camera. Just because an AF sensor has a finite size. Must have. Any camera.

02-12-2011, 03:57 AM   #530
New Member




Join Date: Jan 2011
Posts: 5
Thank you, would you please recommend af chart for me?
02-12-2011, 04:47 AM   #531
Veteran Member
Tommot1965's Avatar

Join Date: Sep 2010
Location: Perth Western Australia
Photos: Gallery | Albums
Posts: 1,026
I use this one



02-12-2011, 04:57 AM   #532
Veteran Member




Join Date: Jan 2009
Location: Budapest
Posts: 821
QuoteOriginally posted by falconeye Quote
Mine uses AF assist only for the first measurement (if at all), not thru hunting. Meaning it would still miss the focus if it wasn't in focus initially. I.e., it does fine focus corrections in the dark even if starting with AF assist light on.
Just went to the dark bathroom to see how mine behaves: approx. 60% of the time the AF assist light blinks for the first measurement and then further adjustments are done in the dark, in 40% of the time it blinks a second time and then further adjustments are still made in the dark and in some cases the cam can't find focus, i.e. after second blink the focus rings moves a bit and I get a blinking "can't lock focus" hexagon - in this situation I'd expect the AF assist light to engage again to help locking focus, seems like the engineers thought it was better to just not focus at all.

Firmware 1.02.21.07
Mfg date: 2010.11.06

02-12-2011, 05:21 AM   #533
Pentaxian




Join Date: Apr 2007
Location: Sweden
Posts: 2,106
QuoteOriginally posted by Tommot1965 Quote
I use this one


I wouldn't recommend that one, it is cluttered with things for the AF to lock on.
02-12-2011, 05:40 AM   #534
New Member




Join Date: Jan 2011
Posts: 5
Would You recommend LensCal from datacolor?
Datacolor SpyderLensCal Press Release
Thank You
02-12-2011, 05:58 AM   #535
Pentaxian
Class A's Avatar

Join Date: Aug 2008
Location: Wellington, New Zealand
Posts: 11,250
QuoteOriginally posted by falconeye Quote
That kind of test chart is not useful at all.
That's not true if it is used as intended.

The idea is that one focuses on the area to the left (what would be left if the image were shown in its correct portrait orientation). In this area, there is only a single black/white transition. This provides the unambiguous target for the AF area.

The part to the right (top in the image) is meant to read out where the focus has been set.

N.B., the focus chart shown is essentially a DIY version of the commercial LensAlign tool (without the alignment aids).


QuoteOriginally posted by Tommot1965 Quote
I use this one
This one only works if the gray print is light enough not to be picked up by the AF module. You need to check whether it can focus on the gray print. If it can, use another version of this chart, with a lighter gray area.

I'm not convinced about this chart because of the uncertainty about whether the gray is "invisible" to the AF module and because the target is actually a bar, rather than a single line/transition. The latter makes it less clear where the camera focused.
02-12-2011, 10:41 AM   #536
Veteran Member
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,871
QuoteOriginally posted by win-win Quote
Thank you, would you please recommend af chart for me?
I use a sheet with two white and two black quadrants, 5° tilted.

It is a bad idea to deduce the amount of front focus or backfocus from looking at some printed area outside the focus point. You can't know if your lens has a flat focus field.

What is worse is that you have to approach the focus chart pretty close to have an empty enough region in the center. I would never calibrate my lens for a target closer than, say 30x the focal length. More about this in a forthcoming paper I am writing as we speak

QuoteOriginally posted by Class A Quote
That's not true if it is used as intended.
You're right. But only very few use it as intended. In the above photo (even if turned into portrait) you see that the image center if far too close to the cluttered area. And that's what I see all the time.

People, pin a sheet of paper with a black line onto a white wall and take a photo perpendicular to it. That's both easiest and most accurate. Just select the calibration setting where the line comes out sharpest. And don't be too close to the wall, a 300mm would need 10m distance ...

Last edited by falconeye; 02-12-2011 at 10:47 AM.
02-12-2011, 11:03 AM   #537
Veteran Member
Smeggypants's Avatar

Join Date: Sep 2010
Posts: 1,536
this is the chart I use. it has single black line in an uncluttered centre area


http://focustestchart.com/focus21.pdf print out page 18


From . Nikon D70 Focus Chart
02-12-2011, 11:42 AM   #538
Veteran Member




Join Date: Oct 2009
Location: United States
Photos: Gallery | Albums
Posts: 392
QuoteOriginally posted by falconeye Quote
I use a sheet with two white and two black quadrants, 5° tilted.
Falk, Can you please elaborate more on your method, and how to measure the blur? I remember you mentioning the method before, that you can exactly measure the blur.

DOes this sheet have alternate black and white quadrants? And does it matter which direction is the tilt, ie, x axis moves clockwise or counter clockwise?

Or, should we wait for your paper?


Thanks,

Abhi
02-12-2011, 11:43 AM   #539
Site Supporter
Site Supporter
GregK8's Avatar

Join Date: May 2008
Location: Western New York
Photos: Gallery | Albums
Posts: 614
QuoteOriginally posted by falconeye Quote
QuoteQuote:
Originally posted by Class A
That's not true if it is used as intended.
You're right. But only very few use it as intended. In the above photo (even if turned into portrait) you see that the image center if far too close to the cluttered area. And that's what I see all the time.
Well, I just tried to use the version with 5% gray, and the K-5 had no problem locking on to the gray area.
02-12-2011, 02:21 PM   #540
Pentaxian
Class A's Avatar

Join Date: Aug 2008
Location: Wellington, New Zealand
Posts: 11,250
QuoteOriginally posted by falconeye Quote
It is a bad idea to deduce the amount of front focus or backfocus from looking at some printed area outside the focus point. You can't know if your lens has a flat focus field.
+1
That's one of the points I make in my AF adjustment hints.

I think the ideal target would provide an unambiguous perpendicular target for focusing and then (e.g., by a turning mechanism), at the very same spot, a tilted plane for the measurement of how big and in which direction the misfocus is.

QuoteOriginally posted by falconeye Quote
More about this in a forthcoming paper I am writing as we speak
Looking forward to reading it.
Too bad that our Pentax cameras aren't programmable. Even better, they could support the whole process with respective AF optimisation functionality.

QuoteOriginally posted by falconeye Quote
In the above photo (even if turned into portrait) you see that the image center if far too close to the cluttered area.
I might be wrong but I thought the photo just shows the apparatus and wasn't a sample of using it. You are correct thought, incorrect attempts of making AF checks/adjustments are legion.

Regarding the distance to the target, doesn't it ultimately depend on which kind of situation (close focusing, infinity, general use) you want the best focus/compromise? I understand it is not always possible to get a setting that works best for all distances so my advice would not be to take a generic distance but think about in which situations focus would be most critical for one personally and than use a corresponding distance during adjustment. It may mean that one arrives at a setting that won't be useful/optimal for general use, though.
Closed Thread

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
camera, dslr, issue, k-5, k-5 ii, k-5 iis, k5, pentax k-5, re, thread
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Help! K7 malfunctioning samski_1 Pentax DSLR Discussion 9 07-11-2010 03:24 PM
Probably not relevant or welcome here climbertrev Canon, Nikon, Sony, and Other Camera Brands 4 05-24-2010 05:15 AM
camera based sensorcleaning how relevant is it. hein Pentax DSLR Discussion 13 09-13-2009 12:26 PM
Light fluff but news.. sort of jeffkrol Pentax News and Rumors 8 12-30-2008 06:30 PM



All times are GMT -7. The time now is 05:01 AM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top