Recently I am worried about the functionality of my Pentax K-50, i.e. I am not able anymore to get real sharp images. Thus my question arose, what I should see as "sharp". So I took several pictures of a non-moving object (my alarm clock) and zoomed in at the focus point. But even there not everything is sharp when I get real close. Why? Should I check my camera, or is this normal? In the example picture, the focus point is exactly between the buttons in the middle.
Example http://imagizer.imageshack.us/v2/150x100q90/540/FWwLaY.jpg

  • \$\begingroup\$ Have you examples of pictures than you find "sharp" taken with the same Pentax ? You can try to take a picture of a ruler inclined at 45° (back focus/front focus problem test). \$\endgroup\$
    – Olivier
    Commented Sep 20, 2014 at 22:20
  • 2
    \$\begingroup\$ The picture you included is too small. Try uploading a full resolution picture. You might also want to take a picture of something with fine detail as your alarm clock probably isn't the best subject for checking sharpness. I usually take pictures of my daughter's eyes, because her eyelashes are very small. \$\endgroup\$ Commented Sep 20, 2014 at 22:31
  • 4
    \$\begingroup\$ There is not enough information to know what is going on. It could be soooo many things or even just your perception. Note that you need to see images at 1:1 (100% magnification) to see the actual pixels, most cameras zoom-in further and interpolate after that which is why it can look blurry without being so. Otherwise, start by checking sharpness settings. \$\endgroup\$
    – Itai
    Commented Sep 20, 2014 at 23:13
  • \$\begingroup\$ photo sharpness is normally disappointing viewed 100%. Which means: dont. \$\endgroup\$ Commented Sep 21, 2014 at 15:47

1 Answer 1


In almost all cases the equipment is just fine and any problems are due to incorrect technique or incorrect camera settings, below are some simple test you can use to see if you are one of the rare cases that really has an equipment fault.

To compere sharpness you should use an image editor that can set to exactly 100% zoom, the camera LCD (or phone, tablet, etc.) is not good enough and many image viewer programs interpolate pixel values when you zoom in too much making the image appear softer.

We are using 100% zoom to be able to compere image sharpness only, never judge you images at 100% because nobody will look at them at that size anyway, also don't compere different cameras at 100% zoom because different pixel counts make sharpness at this zoom level completely irrelevant.

  1. Find an old picture you consider sharp, try to replicate the image and see if you can get the same sharpness - this can give you a good indication if anything changed

  2. Find an object with fine detail, place it in the middle of the frame, put the camera on a tripod, turn off image stabilization, use the lowest ISO, a middle-of-the-road aperture (f/8) and a short shutter speed (this probably means you are taking the image outside in the sunlight), use live view to focus manually (turn on live-view, zoom all the way in, manual focus until you get the sharpest image on screen) - this will give you the sharpest image your camera can take.

  3. Take an image of a flat regular pattern (brick walls are popular) that is exactly parallel the camera's sensor, if some corners are noticeably worse than the others (noticeably- there's always some difference) then something is decentered and you need to get your lens and/or camera recalibrated.

  • \$\begingroup\$ I'm voting this down for this suggestion of setting the viewer to pixel-level zoom. That's usually exactly the wrong thing to do, especially when comparing between cameras. In some cases it might be right for examining the effect of individual parameter changes, but it shouldn't be the baseline. \$\endgroup\$
    – mattdm
    Commented Sep 21, 2014 at 13:20
  • 2
    \$\begingroup\$ Also, the tilted ruler thing is really hard to get right without a careful understanding of how autofocus systems work. \$\endgroup\$
    – mattdm
    Commented Sep 21, 2014 at 13:21
  • \$\begingroup\$ @mattdm - this is not a question about compering cameras, it's about quality degradation in the same camera - and sharpness is easiest to see in 100%, I agree about the 45deg ruler being hard to get right, I'll update the answer \$\endgroup\$
    – Nir
    Commented Sep 21, 2014 at 13:47
  • \$\begingroup\$ It's kind of unclear. There could be a problem — or there could be fear and uncertainty about perfectly fine equipment introduced by over-obsessing about exactly the things you suggest to do. \$\endgroup\$
    – mattdm
    Commented Sep 21, 2014 at 15:21
  • 1
    \$\begingroup\$ The only way the tilted ruler works is to have a high contrast, flat target perpendicular to the optical axis (parallel to the sensor) in the middle of the frame to focus on and the ruler far enough to the side so that the AF doesn't try to focus on a point closer or farther than the point you think it is trying to focus. The actual area covered by a focus point is usually several times the area of the little square in the viewfinder, and the area of highest contrast within the active AF area will be what the camera locks on. See photo.stackexchange.com/a/43792/15871 \$\endgroup\$
    – Michael C
    Commented Sep 21, 2014 at 19:33

Not the answer you're looking for? Browse other questions tagged or ask your own question.