K-x: can't get sharp pictures, I am about to give up

manteiv · 01-20-2010, 02:30 PM

Sorry for all the mess, I am learning how to post pics

I will post real pics tonite of my soft image K-x problem

manteiv · 01-20-2010, 07:11 PM

Here are some pics to illustrate what I meant. On the images from the K-x, it seem that the details are not as sharp as the one I took from the K110d. I need your suggestions to what I can try to improve the details.

The K-x is 12MP, twice as much as the K110D with 6MP so I expect to see a significant improvement in details, which is not the case here.

On the K110D, the images were taken using the 18-55mm DA lens
on the K-x, the images were take with the 18-55mm DAL lens (supposed to be identical to the DA right?)

On the following pics, compare the decorative golden paint on glass. From the K-x image, there is not as much details. On the image from the K110D, you can see the texture of the golden paint very clearly.

Image from K110D:

Image from K-x:

Look at the following jar, the image from the K-x lacks the fine details seen on the one from the K110D

Image from K110D:

Image from K-x:

Damn Brit · 01-20-2010, 07:20 PM

It looks to me like there is a bit of a saturation and contrast issue going on here which could be giving the impression of less than sharp images. Have you tried different settings and (as someone has said) have you uploaded the latest firmware?

dragonfly · 01-20-2010, 07:25 PM

Both examples are taken in different settings which might affect the lighting. Also, there may be different flash strength between the two cameras.

In the first example the orientation of the cups is different, and you're seeing different areas that are in focus.

creampuff · 01-20-2010, 07:32 PM

How can anybody come to a meaningful conclusion about sharpness when the images you post are not from the same angle, distance, metering, focal length or lens? Shoot using one lens on both cameras to rule out variations between the lenses. Shoot in Manual mode to lock in the aperture and shutter speed. Test by focusing manually and using the camera's AF. Use a tripod. Consider the choice of aperture that you're using, shooting at f/16 and f/22 there is a chance of image softening due to diffraction. Sorry but until you remove all the variables, imo your claim about the K-x not being sharp are inconclusive.

Marc Sabatella · 01-20-2010, 07:41 PM

You'll only see more details if you blow the image up bigger than the 6MP camera is capable of. At the posted size, you're not even taking full advantage of the 6MP camera's pixels; the extra pixels of the 12MP camera are just being thrown away.

I notice you have difference sharpness settings on the cameras - try cranking up the sharpness on the K-x to match the K110D. And you're using different lens - no, the DAL version is not identical to the DA. It's identical to the DA II. for the msot part, it's the *better* lens - but that's shooting at "normal" apertures. In these pictures, you're stopping into into the territory where diffraction is really robbing the lenses of their reolution. If you're looking for maximum sharpness. shoot at f/8 or f/11, not f/16 or f/22.

So you might want to repeat the tests with these factors in mind.

**Ash** · 01-20-2010, 09:06 PM

I'm afraid the above reservations hold quite firmly in your posted image examples mantiev.
Different lighting settings, no indication of settings - all quite suspect to be honest.
Furthermore, with the conditions you took these shots in they do appear reasonably acceptable, at this magnfication at least...

More controlled testing is required before any conclusions about your K-x can be made.

manteiv · 01-21-2010, 06:32 AM

thanks for expressing all your concerns.

Unfortunately, I sold my K110D and the stock 18-55mm DA lens so I cannot re-create the comparisons between the images produced by the two cameras.

However, I can remember clearly that I was able to zoom and crop the images from the K110D and see lots of fine details. I have yet to be able to do the same with my K-x, no matter how much I vary the settings, I simply cannot produced images where fine details can be seen sharply when highly magnified.

From the images I posted above, the images of the Greek jar summarize well what I have been experiencing. Although the images from the two cameras were not shot under identical conditions, they are shot under similar conditions. The speed, aperture, ISO, subject and distance are the same so I expect to see the texture and asperity on the surface of the jar quite clearly. The texture on the image from the K-x cannot be seen as clearly.

Maybe it is the limitation of the current lens I am using (Stock 18-55 DAL that came with the K-x). I ordered a 50mm M lens that I will try soon to compare.

I also updated the firmware to 1.01 since day 1 so I don't think firmware is the issue.

In my shootings, I have also noticed that on most images, I obtain better results if I turn OFF the Shake reduction. Maybe that is because I am holding the camera steady on my hands and SR is overcompensating.

Alfisti · 01-21-2010, 07:49 AM

Just do us a favour and post a 100% of images you are unhappy with from the Kx, the EXIF must be there.

Nothing at f/16, f/22 etc .... f/8 please.

manteiv · 01-21-2010, 08:04 AM

Tonite, I am going to shoot a new set of pics at F8 so you can continue to help me out.

Should I boost the sharpness to +1, +2, +3 or leave it at neutral?

By the way, why not at F16 or F22 ? I have always thought that small aperture gives the most depth of field, produce sharper images and can solve focus problems

**Rondec** · 01-21-2010, 08:32 AM

Read a little bit about diffraction limits here. Basically, on an APS-C size sensor, shooting at over f8 will tend to make your photos softer, even if there is an increased depth of field. It is not an absolute limit, just a relative one, but for most lenses, f8 tends to be the sweet spot.

Finger · 01-21-2010, 08:39 AM

Try live view focusing in contrast mode, that's what I do when I want to test the auto focus. If both modes give me the same picture, I know i'm being picky.
It's too bad the DAL lens do not have the distance chart printed on, because it helped me to identify front or back focusing with another lens, there is sometime a big difference under incandescent light.

jeffrey r · 01-21-2010, 11:29 AM

Originally posted by manteiv

Tonite, I am going to shoot a new set of pics at F8 so you can continue to help me out.

Should I boost the sharpness to +1, +2, +3 or leave it at neutral?

By the way, why not at F16 or F22 ? I have always thought that small aperture gives the most depth of field, produce sharper images and can solve focus problems

Maybe try sharpness of +1 or +2. +3 may be overdoing it a bit.

As to the aperture, there should be very few instances where you need to be shooting at apertures of F16, F22 or the like. People usually want "faster" lenses, as in 2, 2.8, etc. Even in those instances, you will usually get sharper pictures between F4-F8. If you generally shoot in the aperture ranges you describe above, try to adjust your technique to bring it closer to the 5.6-8 range for general usage (taking into account exceptions of course).

rawr · 01-21-2010, 12:04 PM

Originally posted by manteiv

...By the way, why not at F16 or F22 ? I have always thought that small aperture gives the most depth of field, produce sharper images and can solve focus problems

Small apertures like f16/f22 etc may indeed produce the greatest depth of field and therefore solve SOME focus/ 'sharpness' problems.

But these sort of apertures, as others have mentioned due to diffraction, actually worsen the actual ability of the lens to resolve detail.

To find out more precisely the sharpest possible settings for your lenses, go to a site like photozone.de and look at their Pentax lens test reports.

Pentax Lens Tests

Their tests will tell you which combination of aperture and focal length will be most likely to deliver you the highest possible resolution out of your lens.

For example, if we look at the review of the SMC-DA 18-55mm f/3.5-5.6 AL at:

Pentax SMC-DA 18-55mm f/3.5-5.6 AL - Review / Test Report

on the second page of their review is a set of tables that show that particular lens, in testing, delivered the highest resolution at a focal length of 35mm at f8, and that in general the highest resolution was achieved at apertures f5.6 and f8, as jeffrey r also mentions in his reply.

Similar results are available for the more recent series II version of this lens (which would be a close match to the 18-55 DAL on the K-x) here:

Pentax smc DA 18-55 mm f/3.5-5.6 AL II review - Image resolution - Lenstip.com

confirming f5.6 to f11 delivers the highest resolution for this lens family.

So in retaking your photos, take your test pix within those f stop ranges if you want to see the maximum resolution of your camera(s) and lenses.

**Ash** · 01-21-2010, 12:07 PM

The limitations of the kit lens will become apparent when shooting above f/11 (diffraction softness) or below f/5.6 (optical softness). If it was the same kit lens you used with the K110D (as you mentioned earlier) there should be no problems getting at least the same detail.

Obviously 100% crops of the 6Mp shot will be at a different magnification as 100% crops of the 12Mp shot, so bear that in mind and make comparisons at 100% for the 6Mp shot and approx 50% of the 12Mp shot as well - just to see the results at the same magnification, before going pixel peeping at 100%.

Edit: I see rawr has summarised this pretty well above...

01-20-2010, 07:11 PM	#17
manteiv Junior Member Join Date: Dec 2009 Location: Montreal Canada Posts: 39 Original Poster	Here are some sample pictures Here are some pics to illustrate what I meant. On the images from the K-x, it seem that the details are not as sharp as the one I took from the K110d. I need your suggestions to what I can try to improve the details. The K-x is 12MP, twice as much as the K110D with 6MP so I expect to see a significant improvement in details, which is not the case here. On the K110D, the images were taken using the 18-55mm DA lens on the K-x, the images were take with the 18-55mm DAL lens (supposed to be identical to the DA right?) On the following pics, compare the decorative golden paint on glass. From the K-x image, there is not as much details. On the image from the K110D, you can see the texture of the golden paint very clearly. Image from K110D: Image from K-x: Look at the following jar, the image from the K-x lacks the fine details seen on the one from the K110D Image from K110D: Image from K-x:

01-21-2010, 06:32 AM	#23
manteiv Junior Member Join Date: Dec 2009 Location: Montreal Canada Posts: 39 Original Poster	thanks for expressing all your concerns. Unfortunately, I sold my K110D and the stock 18-55mm DA lens so I cannot re-create the comparisons between the images produced by the two cameras. However, I can remember clearly that I was able to zoom and crop the images from the K110D and see lots of fine details. I have yet to be able to do the same with my K-x, no matter how much I vary the settings, I simply cannot produced images where fine details can be seen sharply when highly magnified. From the images I posted above, the images of the Greek jar summarize well what I have been experiencing. Although the images from the two cameras were not shot under identical conditions, they are shot under similar conditions. The speed, aperture, ISO, subject and distance are the same so I expect to see the texture and asperity on the surface of the jar quite clearly. The texture on the image from the K-x cannot be seen as clearly. Maybe it is the limitation of the current lens I am using (Stock 18-55 DAL that came with the K-x). I ordered a 50mm M lens that I will try soon to compare. I also updated the firmware to 1.01 since day 1 so I don't think firmware is the issue. In my shootings, I have also noticed that on most images, I obtain better results if I turn OFF the Shake reduction. Maybe that is because I am holding the camera steady on my hands and SR is overcompensating. Last edited by manteiv; 01-21-2010 at 06:41 AM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
K10D - I can't get sharp pictures. Please help!	byrnsy	Troubleshooting and Beginner Help	24	02-09-2010 01:21 AM
When Will He Give it Up?	Wildnsyko	General Talk	16	08-21-2009 07:48 PM
How to take pictures that are sharp throughout ?	pcarfan	Photographic Technique	17	12-30-2008 10:36 AM
I Give Up!!	Fl_Gulfer	Post Your Photos!	14	06-22-2008 06:47 PM
Which menu settings will give me good pictures?	Ole	Pentax DSLR and Camera Articles	5	07-01-2007 06:31 PM

01-20-2010, 02:30 PM	#16
manteiv Junior Member Join Date: Dec 2009 Location: Montreal Canada Posts: 39 Original Poster	Sorry for all the mess, I am learning how to post pics I will post real pics tonite of my soft image K-x problem

01-20-2010, 07:20 PM	#18
Damn Brit Guest	It looks to me like there is a bit of a saturation and contrast issue going on here which could be giving the impression of less than sharp images. Have you tried different settings and (as someone has said) have you uploaded the latest firmware?

01-20-2010, 07:25 PM	#19
dragonfly Veteran Member Join Date: Jun 2009 Location: on the wall Photos: Gallery Posts: 715	Both examples are taken in different settings which might affect the lighting. Also, there may be different flash strength between the two cameras. In the first example the orientation of the cups is different, and you're seeing different areas that are in focus.

01-20-2010, 07:32 PM	#20
creampuff Veteran Member Join Date: May 2007 Location: Singapore Photos: Gallery Posts: 3,953	How can anybody come to a meaningful conclusion about sharpness when the images you post are not from the same angle, distance, metering, focal length or lens? Shoot using one lens on both cameras to rule out variations between the lenses. Shoot in Manual mode to lock in the aperture and shutter speed. Test by focusing manually and using the camera's AF. Use a tripod. Consider the choice of aperture that you're using, shooting at f/16 and f/22 there is a chance of image softening due to diffraction. Sorry but until you remove all the variables, imo your claim about the K-x not being sharp are inconclusive.

01-20-2010, 07:41 PM	#21
Marc Sabatella Veteran Member Join Date: Sep 2006 Location: Denver, CO Photos: Gallery Posts: 10,685	You'll only see more details if you blow the image up bigger than the 6MP camera is capable of. At the posted size, you're not even taking full advantage of the 6MP camera's pixels; the extra pixels of the 12MP camera are just being thrown away. I notice you have difference sharpness settings on the cameras - try cranking up the sharpness on the K-x to match the K110D. And you're using different lens - no, the DAL version is not identical to the DA. It's identical to the DA II. for the msot part, it's the better lens - but that's shooting at "normal" apertures. In these pictures, you're stopping into into the territory where diffraction is really robbing the lenses of their reolution. If you're looking for maximum sharpness. shoot at f/8 or f/11, not f/16 or f/22. So you might want to repeat the tests with these factors in mind.

01-20-2010, 09:06 PM	#22
Ash Loyal Site Supporter Join Date: Mar 2007 Location: Toowoomba, Queensland Photos: Gallery \| Albums Posts: 23,920	I'm afraid the above reservations hold quite firmly in your posted image examples mantiev. Different lighting settings, no indication of settings - all quite suspect to be honest. Furthermore, with the conditions you took these shots in they do appear reasonably acceptable, at this magnfication at least... More controlled testing is required before any conclusions about your K-x can be made.

01-21-2010, 07:49 AM	#24
Alfisti Veteran Member Join Date: Mar 2007 Location: Toronto (for now) Posts: 1,748	Just do us a favour and post a 100% of images you are unhappy with from the Kx, the EXIF must be there. Nothing at f/16, f/22 etc .... f/8 please.

01-21-2010, 08:04 AM	#25
manteiv Junior Member Join Date: Dec 2009 Location: Montreal Canada Posts: 39 Original Poster	Tonite, I am going to shoot a new set of pics at F8 so you can continue to help me out. Should I boost the sharpness to +1, +2, +3 or leave it at neutral? By the way, why not at F16 or F22 ? I have always thought that small aperture gives the most depth of field, produce sharper images and can solve focus problems

01-21-2010, 08:32 AM	#26
Rondec Loyal Site Supporter Join Date: Mar 2009 Location: Gladys, Virginia Photos: Gallery Posts: 27,650	Read a little bit about diffraction limits here. Basically, on an APS-C size sensor, shooting at over f8 will tend to make your photos softer, even if there is an increased depth of field. It is not an absolute limit, just a relative one, but for most lenses, f8 tends to be the sweet spot.

01-21-2010, 08:39 AM	#27
Finger Senior Member Join Date: Mar 2008 Location: Québec Posts: 121	Try live view focusing in contrast mode, that's what I do when I want to test the auto focus. If both modes give me the same picture, I know i'm being picky. It's too bad the DAL lens do not have the distance chart printed on, because it helped me to identify front or back focusing with another lens, there is sometime a big difference under incandescent light.

Thread Tools	Search this Thread
	Search this Thread: Advanced Search

01-21-2010, 11:29 AM	#28
jeffrey r Veteran Member Join Date: Sep 2008 Location: New Jersey, USA Posts: 337	Originally posted by manteiv Tonite, I am going to shoot a new set of pics at F8 so you can continue to help me out. Should I boost the sharpness to +1, +2, +3 or leave it at neutral? By the way, why not at F16 or F22 ? I have always thought that small aperture gives the most depth of field, produce sharper images and can solve focus problems Maybe try sharpness of +1 or +2. +3 may be overdoing it a bit. As to the aperture, there should be very few instances where you need to be shooting at apertures of F16, F22 or the like. People usually want "faster" lenses, as in 2, 2.8, etc. Even in those instances, you will usually get sharper pictures between F4-F8. If you generally shoot in the aperture ranges you describe above, try to adjust your technique to bring it closer to the 5.6-8 range for general usage (taking into account exceptions of course).

01-21-2010, 12:04 PM	#29
rawr Veteran Member Join Date: Jun 2009 Posts: 11,913	Originally posted by manteiv ...By the way, why not at F16 or F22 ? I have always thought that small aperture gives the most depth of field, produce sharper images and can solve focus problems Small apertures like f16/f22 etc may indeed produce the greatest depth of field and therefore solve SOME focus/ 'sharpness' problems. But these sort of apertures, as others have mentioned due to diffraction, actually worsen the actual ability of the lens to resolve detail. To find out more precisely the sharpest possible settings for your lenses, go to a site like photozone.de and look at their Pentax lens test reports. Pentax Lens Tests Their tests will tell you which combination of aperture and focal length will be most likely to deliver you the highest possible resolution out of your lens. For example, if we look at the review of the SMC-DA 18-55mm f/3.5-5.6 AL at: Pentax SMC-DA 18-55mm f/3.5-5.6 AL - Review / Test Report on the second page of their review is a set of tables that show that particular lens, in testing, delivered the highest resolution at a focal length of 35mm at f8, and that in general the highest resolution was achieved at apertures f5.6 and f8, as jeffrey r also mentions in his reply. Similar results are available for the more recent series II version of this lens (which would be a close match to the 18-55 DAL on the K-x) here: Pentax smc DA 18-55 mm f/3.5-5.6 AL II review - Image resolution - Lenstip.com confirming f5.6 to f11 delivers the highest resolution for this lens family. So in retaking your photos, take your test pix within those f stop ranges if you want to see the maximum resolution of your camera(s) and lenses.

01-21-2010, 12:07 PM	#30
Ash Loyal Site Supporter Join Date: Mar 2007 Location: Toowoomba, Queensland Photos: Gallery \| Albums Posts: 23,920	The limitations of the kit lens will become apparent when shooting above f/11 (diffraction softness) or below f/5.6 (optical softness). If it was the same kit lens you used with the K110D (as you mentioned earlier) there should be no problems getting at least the same detail. Obviously 100% crops of the 6Mp shot will be at a different magnification as 100% crops of the 12Mp shot, so bear that in mind and make comparisons at 100% for the 6Mp shot and approx 50% of the 12Mp shot as well - just to see the results at the same magnification, before going pixel peeping at 100%. Edit: I see rawr has summarised this pretty well above...