K-5 motion blur from IS?

Jeff Charles · 06-03-2013, 06:01 AM

I've been getting some blurry photos with my K-5. I've post some examples below with 100% crops to get your ideas about the possible cause.

IS was turned on in each case. I am pretty sure, but not certain, that the IS had "locked" before I took each shot. I also feel that I was pretty well braced each time. The shot of the calendar was taken at 135mm and 1/10 second, so it was at the limit for the IS, but I took several other shots after that one and they were much better.

The photos were to try out the camera after its third repair by C.R.I.S. since November. Does it look like it's still in need of service?

Attached Images

View Picture EXIF

View Picture EXIF

Eulogy · 06-03-2013, 08:56 AM

All those pictures are inherently soft, not due to the shake reduction, but due to the lens. The 18-135 is a great superzoom, but it's still a superzoom; Meaning it's not the sharpest wide open (Which it was, for all these shots). The last two shots are most certainly caused by camera shake, though, I don't really anything having to do with the SR system. (Most people can't reliably freehand shots below 1/30s, 1/10s is hard for even the most stable set of hands)

Try shooting some test photos nonetheless - SR on and off. If the ones with SR disabled are less resolved, you'll know for certain it's miscalibrated.

percy · 06-03-2013, 09:03 AM

If you are using Windows, try looking at the exif data using "ExifToolGUI" (google will find it).

If I recall correctly, I think you will need to download and install "ExifTool" first, and then "ExifToolGUI".

I had a similar issue, and that will tell you whether or not the SR has stabilised.

**dms** · 06-03-2013, 09:30 AM

Seems to be pretty clearly motion--but whether it is camera bounce associated with mirror, or simply camera motion due to slow speed with long FL, or both, and how effective is the camera in controlling same, is something you need to work out by taking photo's with and w/o tripod, a bunch hand held w/ and w/o SR, tripod w/ 2 sec timer, etc.

But for most people 135mm at 1/10 sec is going to be blurred. Why would you test it with such? The rule of thumb is 1/FL (s) for full frame 35mm (if you are fairly secure in your stance). Or 1/135s on 35mm, or 1/200 s on K5. If SR buys you about a couple stops--that's 1/50 sec. And if you ask you really should have worked out/know if you gave SR eneough time. It's easy enough to do anywhere!

BY the way--shoot a burst using SR--if it works, even if your stance is not so good, iikely one or two will be considerably better.

Jeff Charles · 06-03-2013, 09:53 AM

Originally posted by Eulogy

All those pictures are inherently soft, not due to the shake reduction, but due to the lens. The 18-135 is a great superzoom, but it's still a superzoom; Meaning it's not the sharpest wide open (Which it was, for all these shots). The last two shots are most certainly caused by camera shake, though, I don't really anything having to do with the SR system. (Most people can't reliably freehand shots below 1/30s, 1/10s is hard for even the most stable set of hands)

Try shooting some test photos nonetheless - SR on and off. If the ones with SR disabled are less resolved, you'll know for certain it's miscalibrated.

Thanks for your response, but I do not believe that the softness is due to the lens. The 18-135 is much sharper than my sample photos, even wide open at full tele, and especially center frame.

The 1/10 second shots I posted were selected from a number of others that showed little motion blur. In fact, I tried to induce shake by holding the camera with one hand at arm's length, and I could not reproduce the extreme motion blur.

DeadJohn · 06-03-2013, 10:07 AM

The indoor test shots clearly show vertical blur. 1/10 second at 135mm is beyond the limits of SR but I would expect to see horizontal blur, too, whether SR was on or off.

Retry the indoor test on a tripod with SR off to rule out mirror bounce and shutter problems. If things are good on a tripod then try more handheld shots, with SR on and off to see if you find a pattern that triggers the vertical blur.

Include portrait in addition to landscape during the test sequence. When the camera is sideways does the vertical blur turn into horizontal blur? In other words, is the blur direction relative to the camera or to gravity? Mirror and shutter problems will rotate with the camera.

Jeff Charles · 06-03-2013, 10:19 AM

Originally posted by dms

Seems to be pretty clearly motion--but whether it is camera bounce associated with mirror, or simply camera motion due to slow speed with long FL, or both, and how effective is the camera in controlling same, is something you need to work out by taking photo's with and w/o tripod, a bunch hand held w/ and w/o SR, tripod w/ 2 sec timer, etc.

But for most people 135mm at 1/10 sec is going to be blurred. Why would you test it with such? The rule of thumb is 1/FL (s) for full frame 35mm (if you are fairly secure in your stance). Or 1/135s on 35mm, or 1/200 s on K5. If SR buys you about a couple stops--that's 1/50 sec. And if you ask you really should have worked out/know if you gave SR eneough time. It's easy enough to do anywhere!

BY the way--shoot a burst using SR--if it works, even if your stance is not so good, iikely one or two will be considerably better.

The two outdoor shots were not at 1/10, and yet they are blurry. Also, I took a series of the calendar, arms braced on the table, and only a few were as dramatically motion blurred as the one I posted.

My reason for shooting at 1/10 was that I was fooling around with the camera to test it after getting it came back from its third trip to CRIS since November. Also, I do understand the general rule about focal length and shutter speed, but for the calendar shot, I was firmly braced.

My interest in posting this "problem" is to determine if my K-5 is malfunctioning again.

Jeff Charles · 06-03-2013, 10:27 AM

Originally posted by DeadJohn

The indoor test shots clearly show vertical blur. 1/10 second at 135mm is beyond the limits of SR but I would expect to see horizontal blur, too, whether SR was on or off.

Retry the indoor test on a tripod with SR off to rule out mirror bounce and shutter problems. If things are good on a tripod then try more handheld shots, with SR on and off to see if you find a pattern that triggers the vertical blur.

Include portrait in addition to landscape during the test sequence. When the camera is sideways does the vertical blur turn into horizontal blur? In other words, is the blur direction relative to the camera or to gravity? Mirror and shutter problems will rotate with the camera.

Good suggestions. Thanks. I have tried to make the problem happen, but so far, without success. It seems to occur at random. So far, it has not happened with IS off, so that's one data point.

**dms** · 06-03-2013, 11:05 AM

By IS do you mean Shake Reduction (SR)? Presume so.

I recall hearing that motion blur--allegedly do to the camera (lens based) SR--had spoiled an in the jungle jaguar shot. (Years ago.)

If not allowing SR to come on fully could result in more blur than if it was off--it could be an important question in certain types of shooting, such as the aforementioned. But to answer your question--how to decide if it was camera malfunction or not would be tough--except over longer time with extensive experience. But from your later discussion it sounds like (in all probability) SR just hadn't come on--and thus the blur was to be expected.

I often wonder if Pentax couldn't have installed a time delay--so once activated the SR remains on for a period of time (e.g., 30 s). I often forget to keep the shutter release half pushed when I am anticipating the critical moment.

**Lowell Goudge** · 06-03-2013, 11:36 AM

What I find investing is that the first two or so shots are relatively sharp, what you would expect from the conditions , but the last ones all seem to have a lot of vertical only blurr. Are these done on a tripod? Or some other vertical support?

percy · 06-03-2013, 11:58 PM

Jeff,
Have you checked the exif data to see whether or not the SR had stabilised on the shots in question?

Jeff Charles · 06-04-2013, 03:45 AM

Originally posted by percy

Jeff,
Have you checked the exif data to see whether or not the SR had stabilised on the shots in question?

I checked using Exiftool. SR had stabilized on all four of the photos I posted.

aurele · 06-04-2013, 03:56 AM

to me, it's just that the SR can't do everything

There is nothing malfunctionning here, you just need a faster shutter speed.

And i would suggest you to disable SR in daylight and keep the shutter @ 1/ focal lenght * 1.5 sec to make sure shots are crisps.

Jeff Charles · 06-04-2013, 04:00 AM

Originally posted by Lowell Goudge

What I find investing is that the first two or so shots are relatively sharp, what you would expect from the conditions , but the last ones all seem to have a lot of vertical only blurr. Are these done on a tripod? Or some other vertical support?

The first two shots are not close to as sharp as many others I've taken with the 18-135 under similar conditions. I did not use a tripod for any of the photos. For the calendar photo, my elbows were braced on a table.

As I mentioned in another post, I've tried unsuccessfully to cause the problem by setting the lens to full zoom, the camera to 1/10 second, and the SR on, and then holding the camera at arms length with one hand to take the photo.

Jeff Charles · 06-04-2013, 04:27 AM

Originally posted by aurele

to me, it's just that the SR can't do everything

There is nothing malfunctionning here, you just need a faster shutter speed.

And i would suggest you to disable SR in daylight and keep the shutter @ 1/ focal lenght * 1.5 sec to make sure shots are crisps.

The barn photo was taken at 1/200, with a focal length of 28mm. The car photo was 1/400 and 48mm. Even without SR, those are more than adequate shutter speeds.

The calendar shot was at 1/10 second, and at full zoom, so I agree that it was pushing the limits of SR. However, I took a number of other photos of the calendar using the same settings, and the SR compensated reasonably well. The problem has also occurred other times, seemingly at random. I haven't detected a pattern.

Also, this behavior is new. I used the camera for a year up until last November, and I do not recall similar results. Since then, as I mentioned in an earlier post, the camera has been to C.R.I.S. three times for separate repairs. If it has to go back to them again, and that's the way I am leaning, I am going to ask Pentax to replace it.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Depicting motion: Panning and motion blur	slackercruster	Mini-Challenges, Games, and Photo Stories	109	09-07-2021 09:59 AM
Trouble with motion blur	Crust	Pentax K-01	10	02-11-2013 04:02 PM
k-7 and fill flash gives me shake/blur/motion images?	netuser	Troubleshooting and Beginner Help	5	07-11-2010 03:22 AM
Motion Blur Portrait	jmbradd	Post Your Photos!	2	03-11-2009 01:30 PM
Motion blur	Donald	Post Your Photos!	3	04-30-2007 07:23 PM

06-03-2013, 08:56 AM	#2
Eulogy Veteran Member Join Date: Feb 2012 Location: Fredericton, New Brunswick Photos: Gallery Posts: 632	All those pictures are inherently soft, not due to the shake reduction, but due to the lens. The 18-135 is a great superzoom, but it's still a superzoom; Meaning it's not the sharpest wide open (Which it was, for all these shots). The last two shots are most certainly caused by camera shake, though, I don't really anything having to do with the SR system. (Most people can't reliably freehand shots below 1/30s, 1/10s is hard for even the most stable set of hands) Try shooting some test photos nonetheless - SR on and off. If the ones with SR disabled are less resolved, you'll know for certain it's miscalibrated.

06-03-2013, 09:03 AM	#3
percy Senior Member Join Date: May 2011 Location: Herefordshire, UK Posts: 197	If you are using Windows, try looking at the exif data using "ExifToolGUI" (google will find it). If I recall correctly, I think you will need to download and install "ExifTool" first, and then "ExifToolGUI". I had a similar issue, and that will tell you whether or not the SR has stabilised.

06-03-2013, 09:30 AM	#4
dms Site Supporter Join Date: Aug 2011 Location: New York, NY Photos: Gallery Posts: 2,192	Seems to be pretty clearly motion--but whether it is camera bounce associated with mirror, or simply camera motion due to slow speed with long FL, or both, and how effective is the camera in controlling same, is something you need to work out by taking photo's with and w/o tripod, a bunch hand held w/ and w/o SR, tripod w/ 2 sec timer, etc. But for most people 135mm at 1/10 sec is going to be blurred. Why would you test it with such? The rule of thumb is 1/FL (s) for full frame 35mm (if you are fairly secure in your stance). Or 1/135s on 35mm, or 1/200 s on K5. If SR buys you about a couple stops--that's 1/50 sec. And if you ask you really should have worked out/know if you gave SR eneough time. It's easy enough to do anywhere! BY the way--shoot a burst using SR--if it works, even if your stance is not so good, iikely one or two will be considerably better.

06-03-2013, 09:53 AM	#5
Jeff Charles Senior Member Join Date: Nov 2006 Location: Oaxaca, Mexico; Shohola PA, USA Posts: 219 Original Poster	Originally posted by Eulogy All those pictures are inherently soft, not due to the shake reduction, but due to the lens. The 18-135 is a great superzoom, but it's still a superzoom; Meaning it's not the sharpest wide open (Which it was, for all these shots). The last two shots are most certainly caused by camera shake, though, I don't really anything having to do with the SR system. (Most people can't reliably freehand shots below 1/30s, 1/10s is hard for even the most stable set of hands) Try shooting some test photos nonetheless - SR on and off. If the ones with SR disabled are less resolved, you'll know for certain it's miscalibrated. Thanks for your response, but I do not believe that the softness is due to the lens. The 18-135 is much sharper than my sample photos, even wide open at full tele, and especially center frame. The 1/10 second shots I posted were selected from a number of others that showed little motion blur. In fact, I tried to induce shake by holding the camera with one hand at arm's length, and I could not reproduce the extreme motion blur.

06-03-2013, 10:07 AM	#6
DeadJohn Pentaxian Join Date: Jan 2011 Location: New York Posts: 4,834	The indoor test shots clearly show vertical blur. 1/10 second at 135mm is beyond the limits of SR but I would expect to see horizontal blur, too, whether SR was on or off. Retry the indoor test on a tripod with SR off to rule out mirror bounce and shutter problems. If things are good on a tripod then try more handheld shots, with SR on and off to see if you find a pattern that triggers the vertical blur. Include portrait in addition to landscape during the test sequence. When the camera is sideways does the vertical blur turn into horizontal blur? In other words, is the blur direction relative to the camera or to gravity? Mirror and shutter problems will rotate with the camera.

06-03-2013, 10:19 AM	#7
Jeff Charles Senior Member Join Date: Nov 2006 Location: Oaxaca, Mexico; Shohola PA, USA Posts: 219 Original Poster	Originally posted by dms Seems to be pretty clearly motion--but whether it is camera bounce associated with mirror, or simply camera motion due to slow speed with long FL, or both, and how effective is the camera in controlling same, is something you need to work out by taking photo's with and w/o tripod, a bunch hand held w/ and w/o SR, tripod w/ 2 sec timer, etc. But for most people 135mm at 1/10 sec is going to be blurred. Why would you test it with such? The rule of thumb is 1/FL (s) for full frame 35mm (if you are fairly secure in your stance). Or 1/135s on 35mm, or 1/200 s on K5. If SR buys you about a couple stops--that's 1/50 sec. And if you ask you really should have worked out/know if you gave SR eneough time. It's easy enough to do anywhere! BY the way--shoot a burst using SR--if it works, even if your stance is not so good, iikely one or two will be considerably better. The two outdoor shots were not at 1/10, and yet they are blurry. Also, I took a series of the calendar, arms braced on the table, and only a few were as dramatically motion blurred as the one I posted. My reason for shooting at 1/10 was that I was fooling around with the camera to test it after getting it came back from its third trip to CRIS since November. Also, I do understand the general rule about focal length and shutter speed, but for the calendar shot, I was firmly braced. My interest in posting this "problem" is to determine if my K-5 is malfunctioning again.

06-03-2013, 10:27 AM	#8
Jeff Charles Senior Member Join Date: Nov 2006 Location: Oaxaca, Mexico; Shohola PA, USA Posts: 219 Original Poster	Originally posted by DeadJohn The indoor test shots clearly show vertical blur. 1/10 second at 135mm is beyond the limits of SR but I would expect to see horizontal blur, too, whether SR was on or off. Retry the indoor test on a tripod with SR off to rule out mirror bounce and shutter problems. If things are good on a tripod then try more handheld shots, with SR on and off to see if you find a pattern that triggers the vertical blur. Include portrait in addition to landscape during the test sequence. When the camera is sideways does the vertical blur turn into horizontal blur? In other words, is the blur direction relative to the camera or to gravity? Mirror and shutter problems will rotate with the camera. Good suggestions. Thanks. I have tried to make the problem happen, but so far, without success. It seems to occur at random. So far, it has not happened with IS off, so that's one data point.

06-03-2013, 11:05 AM	#9
dms Site Supporter Join Date: Aug 2011 Location: New York, NY Photos: Gallery Posts: 2,192	By IS do you mean Shake Reduction (SR)? Presume so. I recall hearing that motion blur--allegedly do to the camera (lens based) SR--had spoiled an in the jungle jaguar shot. (Years ago.) If not allowing SR to come on fully could result in more blur than if it was off--it could be an important question in certain types of shooting, such as the aforementioned. But to answer your question--how to decide if it was camera malfunction or not would be tough--except over longer time with extensive experience. But from your later discussion it sounds like (in all probability) SR just hadn't come on--and thus the blur was to be expected. I often wonder if Pentax couldn't have installed a time delay--so once activated the SR remains on for a period of time (e.g., 30 s). I often forget to keep the shutter release half pushed when I am anticipating the critical moment.

06-03-2013, 11:36 AM	#10
Lowell Goudge Site Supporter Join Date: Jan 2007 Location: Toronto Photos: Gallery \| Albums Posts: 17,888	What I find investing is that the first two or so shots are relatively sharp, what you would expect from the conditions , but the last ones all seem to have a lot of vertical only blurr. Are these done on a tripod? Or some other vertical support?

06-03-2013, 11:58 PM	#11
percy Senior Member Join Date: May 2011 Location: Herefordshire, UK Posts: 197	Jeff, Have you checked the exif data to see whether or not the SR had stabilised on the shots in question?

06-04-2013, 03:45 AM	#12
Jeff Charles Senior Member Join Date: Nov 2006 Location: Oaxaca, Mexico; Shohola PA, USA Posts: 219 Original Poster	Originally posted by percy Jeff, Have you checked the exif data to see whether or not the SR had stabilised on the shots in question? I checked using Exiftool. SR had stabilized on all four of the photos I posted.

06-04-2013, 03:56 AM	#13
aurele Veteran Member Join Date: Jan 2011 Location: Paris, France Photos: Albums Posts: 3,217	to me, it's just that the SR can't do everything There is nothing malfunctionning here, you just need a faster shutter speed. And i would suggest you to disable SR in daylight and keep the shutter @ 1/ focal lenght * 1.5 sec to make sure shots are crisps.

06-04-2013, 04:00 AM	#14
Jeff Charles Senior Member Join Date: Nov 2006 Location: Oaxaca, Mexico; Shohola PA, USA Posts: 219 Original Poster	Originally posted by Lowell Goudge What I find investing is that the first two or so shots are relatively sharp, what you would expect from the conditions , but the last ones all seem to have a lot of vertical only blurr. Are these done on a tripod? Or some other vertical support? The first two shots are not close to as sharp as many others I've taken with the 18-135 under similar conditions. I did not use a tripod for any of the photos. For the calendar photo, my elbows were braced on a table. As I mentioned in another post, I've tried unsuccessfully to cause the problem by setting the lens to full zoom, the camera to 1/10 second, and the SR on, and then holding the camera at arms length with one hand to take the photo.

06-04-2013, 04:27 AM	#15
Jeff Charles Senior Member Join Date: Nov 2006 Location: Oaxaca, Mexico; Shohola PA, USA Posts: 219 Original Poster	Originally posted by aurele to me, it's just that the SR can't do everything There is nothing malfunctionning here, you just need a faster shutter speed. And i would suggest you to disable SR in daylight and keep the shutter @ 1/ focal lenght * 1.5 sec to make sure shots are crisps. The barn photo was taken at 1/200, with a focal length of 28mm. The car photo was 1/400 and 48mm. Even without SR, those are more than adequate shutter speeds. The calendar shot was at 1/10 second, and at full zoom, so I agree that it was pushing the limits of SR. However, I took a number of other photos of the calendar using the same settings, and the SR compensated reasonably well. The problem has also occurred other times, seemingly at random. I haven't detected a pattern. Also, this behavior is new. I used the camera for a year up until last November, and I do not recall similar results. Since then, as I mentioned in an earlier post, the camera has been to C.R.I.S. three times for separate repairs. If it has to go back to them again, and that's the way I am leaning, I am going to ask Pentax to replace it.