Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
10-20-2015, 01:44 PM   #1
Junior Member




Join Date: Nov 2012
Posts: 38
Focus test feedback

Hi all, I've been testing focus a bit today with some shots inside the apt. I would like to confirm something.

So the first picture is the overall shot, and the second is a 100% crop. Shot on the DA 50mm f1.8, using AFS, SEL1 center point.

I apologize if this is a noob question, but I cannot get the K-3 to focus on the cable box on top of the table. I put my focus point on the gray area, with a hint of overlap with the vents closest to me. However, this will happen even when the focus point is squarely on top of those vents.

What would be the assessment to be made here? I feel like there is enough contrast that it should be able to focus accurately. I know the lens does focus accurately in other situations so it does not back or front focus otherwise.

Thanks in advance.

Attached Images
View Picture EXIF
PENTAX K-3  Photo 
View Picture EXIF
PENTAX K-3  Photo 
10-20-2015, 01:59 PM   #2
Site Supporter




Join Date: Jan 2011
Photos: Albums
Posts: 2,237
It looks like the focus is at the very end of the cable box. You are facing the window, that might have something to do with it.
10-20-2015, 02:05 PM   #3
Junior Member




Join Date: Nov 2012
Posts: 38
Original Poster
Yes it appears that the focus always ends up being past it, thought the focus point is not above the closer vents. I guess I will test with the lighting in front, but would appreciate more input as well.
10-20-2015, 03:58 PM   #4
Veteran Member




Join Date: Mar 2012
Location: Perrineville, NJ
Posts: 1,364
The focus indicators in the viewfinder are not the actual focus points, and can be misaligned relative to each other. It's enough that you cannot be sure what the AF sensors are actually focusing on. Also the little LEDs used as the focus indicators are smaller than the actual focus zones. This was a major problem with the K-5 due to its large focus zones. I know your EXIF says you are using the K-3 and its focus points are smaller, but I think they are still larger than the LEDs, and the alignment can still be an issue. You should repeat the test using better focus targets, like staggered books on your desk, to eliminate any chance of missing the focus zone.

10-20-2015, 04:04 PM   #5
Site Supporter
DeepThrob's Avatar

Join Date: Nov 2012
Location: Taos, New Mexico
Posts: 137
Go here:
http://http://petapixel.com/2013/03/12/ghettoca-a-diy-lens-calibration-tool-...enabled-dslrs/

get the GettoCAL pdf template, print it or have Walmart do it. Then you'll have something to focus on,,,
10-20-2015, 05:54 PM   #6
Pentaxian
amoringello's Avatar

Join Date: May 2008
Location: Virginia, USA
Photos: Gallery
Posts: 1,393
Put a black dot (about 1/4 inch) on a white sheet of paper. Put the camera in continual focus.
While focusing, move the sheet of paper around the selected focus spot.
You will be able to get a basic outline where the focus point/area is located and just how large that area is.

On one of my old cameras (K7?), the center focus spot was centered below the lower left off the "( )". (i.e. it was way off)
I could have sent it in for repair and been without it for weeks (and with the repeated experience with CRIS Camera, the repair would have been 12 weeks and not done correctly).
I opted to simply live with knowing where the focus area was centered.
10-20-2015, 08:09 PM   #7
Senior Member




Join Date: Mar 2013
Location: So-central, MI / Central, IN
Posts: 172
I cannot comment specific to the K3, but do have some experience with the K5:

The K5 AF focus performance seemed to be sensitive to the type of lighting, therefore I'd recommend that testing be performed in daylight for typical general photography unless the user has reason to optimize for incandescent lighting.

Ditto the recommendation of the Amoringello- as an exercise to become familiar with the center AF sensor. I used a penny taped to a wall. Note the configuration and location of the sensor relative to the ( ) markings of the view finder. Bear in mind that any most significant contrast that falls across the sensor (e.g. background) is going to steal the attention of the sensor. Therefore your decisions for sensor placement on your subject should always keep this in mind.

Also the recommendation to use a staggered (e.g. stepped) pattern of books might be a good idea. However keep in mind that the subject selection or camera distance should be controlled such that the AF center sensor falls completely with in the single center target subject (e.g. book spine) without overlapping one of its edges. I have found this staggered type of target to be useful at times.

I recommend to use a tripod and a consistent approach of setting focus at a fixed and different setting (e.g. further or closer) between test shots (this can be done manually or using electronic autofocus), then one press AF auto focus on your target for each test shot.

As an alternate strategy a single plane subject can be used if it is set up methodically (e.g. wall or 1/2 of a 4x8 foot plywood board)- restrict the test patterns to black and white (e.g. can be printed from a laser or inkjet printer). Test patterns can be samples from test target graphics and/or text. The patterns should be placed in a symmetrical manner near the corners or edges of the target. Symmetry of the pattern at each corner/edge is important because this will help to compensate for small misalignment in perpendicularity of the camera to target during image evaluation. Equally important is that the center AF sensor target consists of a single prominent black line surrounded by an area of white thus limiting the sensor to find only the single consistent target for each test shot.

AF fine adjustment testing and calibration can be a challenging chore. A good set up with technique that aims at being consistent and repeatable will help. The end results can be very worthwhile. My limited experience with the K5 has been that the consistency of the AF performance is highly dependent on getting an accurate AF fine adjustment. And also being mindful of the sensor size. My autofocus accuracy is very consistent, and hunting is rare except in low light.
10-20-2015, 08:22 PM   #8
Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 26,233
QuoteOriginally posted by thehiko Quote
What would be the assessment to be made here? I feel like there is enough contrast that it should be able to focus accurately.
You put the intended target on a receding plane. I would expect the actual plane of focus to be anywhere +/- 5cm of what you intended. Unfortunately, the camera cannot read your mind and will attain focus on whatever portion of the AF point (the diameter is wider than you think) that provides highest contrast at the moment. Factor in the intrinsic precision (or lack thereof), focus could be anywhere. Do the same test on a flat target parallel to the sensor plane and prepare to be much happier.


Steve

10-21-2015, 09:29 AM   #9
Junior Member




Join Date: Nov 2012
Posts: 38
Original Poster
Thank you so much everyone for all your experiences and suggestions.

I have tested the center focus point per amoringello's procedures, and found that the center point for my K-3 covers almost exactly inside the ( ). Top and bottom coverage is kept within an imaginary straight line drawn from the ends of the parentheses.

I can see now that with the picture I took, the center point focus area actually covers the farther vents on the cable box, even a little bit of the Playstation. So I can understand why it decided to focus there, being that the contrast is greater. Also, because the setup is indeed on a receding plane, minute vertical differences in where the camera decided to focus resulted in a drastic distance change. Consequently, I tested with adequate lighting on the front side, and the focus hit about 90% of the time where I intended it, and not as far off when it did miss. So a backlit scene with not enough front lighting also affected its accuracy.

To summarize,
- a backlit scene with relatively little front lighting
- a wide focus point coverage area
- less relative contrast within the focus area
- subject being on a receding plane
all contributed to not being able to focus where I intended. But I do wish I had my old Nikon with me to compare.

On a side note, while researching to see if this was something that was experienced by other systems, I ran into a 7D Mark ii review and found this snippet sort of interesting (from the-digital-picture.com):

"An apparently undocumented attribute of the outer AF points (at least in single AF point mode) is that they all can use contrast information under the next AF point toward the center, causing the adjacent AF point to functionally appear as a focus assist-like point. When contrast becomes weak (or even non-existent) on a selected outer AF point and the next point down/up (for top/bottom AF points) or left/right (for right/left AF points) has good contrast, the 7D II will auto focus using the position of that particular assist point."

Of course the 7D Mark ii has one of the best AF in the business so I'm not directly comparing, but in essence, it sounds like this is happening to me but just within the focus point because of the coverage size.
10-21-2015, 08:28 PM   #10
Senior Member




Join Date: Mar 2013
Location: So-central, MI / Central, IN
Posts: 172
Glad to hear you're making progress with evaluating your experience to date. I think you will find that the knowledge gained will be very helpful and I trust that you shall find you get very good AF results with the K3 in the future. Although smaller AF sensor sizes would be highly desirable- I've still experienced very good results with my K5 which incidentally measures as you described for the center sensor.
10-22-2015, 05:44 AM   #11
Veteran Member




Join Date: Mar 2012
Location: Perrineville, NJ
Posts: 1,364
QuoteOriginally posted by Tanzer Quote
... using better focus targets, like staggered books on your desk, to eliminate any chance of missing the focus zone.
What I meant to say was to use something like hardcover books placed upright on your desk, to present a large flat center focus target. The other books can be staggered at different distances. Well I guess those other books could be turned and viewed "on end" to keep them compact within the frame, since you wouldn't be focusing on them per se, just using them to judge depth of focus. This is hard to explain ... wish I had some kind of hobby where, um, I could express my meaning using some kind of visual communication instead of writing like a thousand words ....
Reply

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!
dslr, focus, k-3, k3, pentax k-3, shot
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Close Focus Test - Pentax K 28 f2.0 vs Sigma 30 f1.4 mannyquinto Pentax SLR Lens Discussion 5 05-02-2014 12:00 AM
Samyang 14mm Focus to Infinity Photo Sample Test Nitrogliserin Troubleshooting and Beginner Help 7 04-14-2014 01:16 AM
Nature Squirrel Test focus Hamilton Photo Critique 4 03-10-2011 01:19 PM
Focus Test Mallee Boy Post Your Photos! 9 08-30-2009 07:40 AM
New Owner & Focus Test mathan Pentax SLR Lens Discussion 2 11-03-2008 01:20 PM



All times are GMT -7. The time now is 09:59 PM. | See also: NikonForums.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