[dgaies]

because LV is using contrast detection to auto focus where VF uses phase detection (unless you set LV to use otherwise). VF should be accurate on the k-r except for the Tungsten light issue.

Should be, I agree. I was just saying the contrast detection seems to be very accurate and isn't affected by FF/BF issues of individual lenses like the traditional phase detection AF.

[gp1806]

I have a question. Can we see some examples?

This was shot in window daylight, no artificial light, P mode, AWB, Center focus point on the cat's head, Optical view finder focus. In my albums, under K-R test image 5824. Notice the carpet in front of the catnip can, it is in focus.



This was the same except Live View focus. In my albums, under K-R test image 5827.




I had to copy and paste the URL to post.

[GerryL]

First go to basics since using a DSLR is different from using an SLR.
For one thing the focus points will always be off if one does not know how to use it or would be better if you just use the center point like in SLR's.
The SR is another thing, half-pressing the shutter release still works with DSLR's but instead of triggering the meter, it focuses on the DSLR, it also engages the SR or Shake Reduction feature which must be "locked" before the shutter is fully tripped.
If AF is disengaged, most specially the SR feature would be easily overlooked by the operator. You lose the SR feature and incur blur instead.
If using a tripod, disengage the SR feature too.
Try to take some photos with the camera to familiarize yourself fully with its features and quirks. It's like buying a new car, by driving it always you will find that you will find ways to make it drive better..or find faults that you must compensate upon.

[GerryL]

This was shot in window daylight, no artificial light, P mode, AWB, Center focus point on the cat's head, Optical view finder focus. In my albums, under K-R test image 5824.

This was the same except Live View focus. In my albums, under K-R test image 5827.

Sorry, I can't post pics, I don't know why.

Image size might be too large. Re-size and re-post. There are certain image size limitations if not a "site supporter".

[Bill_R]

Image size might be too large. Re-size and re-post. There are certain image size limitations if not a "site supporter".

Or post a link to the photo on Flickr or Picasa or somewhere. The smaller non-supporter storage might be used up,

[gp1806]

First go to basics since using a DSLR is different from using an SLR.
For one thing the focus points will always be off if one does not know how to use it or would be better if you just use the center point like in SLR's.
The SR is another thing, half-pressing the shutter release still works with DSLR's but instead of triggering the meter, it focuses on the DSLR, it also engages the SR or Shake Reduction feature which must be "locked" before the shutter is fully tripped.
If AF is disengaged, most specially the SR feature would be easily overlooked by the operator. You lose the SR feature and incur blur instead.

I used center point and recompose. I did get the SR signal in the viewfinder. This was a test of auto focus in dim light, but not tungsten light. This type of high ISO shot was why I bought the K-R. My K-R front focuses in dim daylight as well as tungsten light. Thanks for the reply, it's appreciated.

[stormcloud]

center point and recompose might be the issue you've got a pretty narrow depth of field there

still you should do a focus test ona ruler or something and check for ff/bf and adjust for each lens

[hcarvalhoalves]

I used center point and recompose. I did get the SR signal in the viewfinder. This was a test of auto focus in dim light, but not tungsten light. This type of high ISO shot was why I bought the K-R. My K-R front focuses in dim daylight as well as tungsten light. Thanks for the reply, it's appreciated.

From the samples shots, this doesn't look like front focus to me, it's off by a large margin. I guess it's more of user error, it may help getting used to how AF works.

If you used center point and recomposed, something went wrong there. You can have AF-A/AF-C on instead of AF-S, it seems to me it tried to focus again when you released shutter because it's completely off. Have you checked that?

[Bill_R]

When you used CP focus on the cats head did you keep the shutter half pressed when you recomposed?
Way too narrow DOF to catch much in those shots.

[Philoslothical]

There is a lot wrong with the test pic. It's way overexposed, due to your choice of settings:

Exposure Time 1/60s (0.017) Aperture F/1.8 ISO Equivalent 1600

Huge aperture and high ISO, for an indoor daylight shot? Then throw in the 1/60th exposure (which looks very long for these settings). You could have shot this at ISO 200 with that aperture, and probably half the exposure time as well. If you prefer to go with such a high ISO, decrease your aperture to f/8 or so and get the whole cat in focus. Even if the focal plane had intersected the cat's head, that's all that could have been in focus with that aperture and distance (illustrated by your second pic). A slight accidental jerk of the camera when you shot this could produce exactly what we're seeing here. If you didn't have a firm grip, the camera dipping slightly when you go to take the shot is plausible, especially if it was at an awkward angle (it looks like you're pretty close to the floor here).

I'm not trying to be overly critical or discouraging here. I'm still very new to this stuff too. I just think that for a test shot you had everything possible working against you. Perhaps try it again with more moderate settings, and see if the problem persists? I realize that the AF isn't directly tied to the settings you chose, but they couldn't have helped. Also, for testing, a tripod is ideal. It removes much of the possibility of shake or unintended movement, especially if you also use the 2 second self timer.

[gp1806]

From the samples shots, this doesn't look like front focus to me, it's off by a large margin. I guess it's more of user error, it may help getting used to how AF works.

If you used center point and recomposed, something went wrong there. You can have AF-A/AF-C on instead of AF-S, it seems to me it tried to focus again when you released shutter because it's completely off. Have you checked that?

The focus is off by a couple of inches. The carpet right in front of the catnip can is in focus. See the different look of the carpet in this area in both pics. This was a focus test. Camera is set to AFS. I chose a lens that would show focus errors, FA 50mm f/1.4. Is this a bad example to send to Pentax? What sort of pics would be good to send to Pentax?

[gp1806]

When you used CP focus on the cats head did you keep the shutter half pressed when you recomposed?

Yes.

Way too narrow DOF to catch much in those shots.

That was the point. Narrow DOF will show focus errors.

[gp1806]

There is a lot wrong with the test pic. It's way overexposed, due to your choice of settings:

Exposure Time 1/60s (0.017) Aperture F/1.8 ISO Equivalent 1600

Shot in P mode as part of the test. Camera picked settings.

[garyk]

I believe it. I have the same problem. Crazy sometimes it is spot on and with center point focus. Sometimes it just misses... ??? I am going to buy another brand and see if it is me or the camera. Maybe me. But I am going to find out.

[hcarvalhoalves]

The focus is off by a couple of inches. The carpet right in front of the catnip can is in focus. See the different look of the carpet in this area in both pics. This was a focus test. Camera is set to AFS. I chose a lens that would show focus errors, FA 50mm f/1.4. Is this a bad example to send to Pentax? What sort of pics would be good to send to Pentax?

The focus is completely off for a 50mm, but there are other wrong settings for this pic (pretty thin DoF, ISO too high, shooting against light, overexposed) that I don't know if it's trustable to base a conclusion on it.

You should make other experiments, with more standard settings. You might be able to narrow the problem down to a camera body / lens or user error. The second shot - using Live View - is well in focus, so doesn't look like a problem with your particular lens. If you consistently gets out of focus shots, I would expect it being a camera malfunction, but if it's more like 50/50 I doubt so.