K-X Highlight Clipping?

trailrider · 03-03-2010, 06:34 PM

I am new to the forum and considering a K-X purchase. I have read the review at dpreview. I am especially interested in the dynamic range test.
Pentax K-x Review: 17. Photographic tests (DR): Digital Photography Review

Do people feel the following comment is accurate? If so can it be overcome via adjustments in pre and post processing?

"The K-x has a fairly pronounced tendency to clip highlights and over exposed parts of the image are usually not recoverable at all or produce false color information when negative digital exposure compensation is applied in a RAW converter."

I am no expert in photography. I have been studying the dynamic range curves. I see just the K-X being just 1/2 EV lower than say a canon. Even when they use their highlight correction. I wouldn't think that would make that much difference. Is there a way to shift the dynamic range curve to the right? To me it looks like the K-X does a better job of shadow recovery. One of the reasons for me in moving to a DSLR is less highlight clipping.

Thanks,
Mike

er1kksen · 03-03-2010, 06:48 PM

Basically, if the DR is the same, regardless of whether there's more in the highlights or shadows, then the DR is the same. The only difference, really, is where on the curve the camera exposes. YOU have control over this: if you're using a Pentax with smaller highlight range and more in the shadows, expose to preserve the highlights and bring the exposure up to whatever's visually "correct" in post. If you're using a Canon with more range in the highlights and less in the shadows, expose to the right (brighter) to take advantage of that range and then bring the exposure down to whatever's visually "correct" in post. The end result is essentially the same.

Wheatfield · 03-03-2010, 06:51 PM

Remember that these tests tend to be designed to test to failure rather than testing to reality.
You may find that you have to dial in some exposure comp in certain scenes, but it likely won't matter what brand of camera you are using in this regard..

trailrider · 03-03-2010, 08:37 PM

I appreciate your replys. I did not study the conclusion information thoroughly. It states:
"One of the K-x's main weaknesses is its habit to clip highlights. The problem is exacerbated by a tendency to overexpose high contrast scenes. The K-x also offers very little in terms of RAW headroom. Therefore shooting RAW will not solve the problem entirely. However, the ability to adjust the tone curve during file conversion helps to mitigate this issue. Your best bet though is to apply some negative exposure compensation when necessary. "
"
All in all the K-x is perfectly capable of producing high quality image output, especially in low light. For beginners who primarily shoot in Auto mode the tendency to clip highlights can be slightly problematic. For everyone else it's controllable via some manual intervention."

Mike

rawr · 03-04-2010, 05:32 AM

It's not something I have noticed terribly much in my use of the camera, tbh.

If you shoot in RAW you do actually get a lot of headroom with the K-x. I also think the reviewer needs to think through the fact that every RAW convertor brings it's own 'personality' to the conversion process. Whatever he tried, ACR 2.6 and the included Pentax DCU 4 handles the K-x's RAW output pretty well, highlights and all. Even Picasa and others that use the dcraw engine do OK now that they know about the K-x.

While I am not a massive pixel peeper, I still do care about IQ, and I don't see a big issue here.

jlaubza · 03-04-2010, 11:39 AM

Hi, I tested some RAW images with the K-X in Adobe CS4. Highlights were very clipped and in order to recover these, I had to lower the exposure value of the images quite a lot, ending up with dark images. The limited DR of the KX became very obvious. So, if you get this camera, be prepared for clipped highlights - the DR is visibly less than a Canon or Nikon but the noise factor is better than either the 500d or D90 at the same high iso. At lower iso's (up to 1600) they are much the same.

The problem with the K-X is that it has been obviously crippled by cutting out features that would have not made much difference to production cost.

er1kksen · 03-04-2010, 11:57 AM

jlaubza, the very fact that you were able to recover those highlights demonstrates that the K-x does NOT have limited DR (compared to the 500D and D90). What you need is a program like ACR or lightroom with a "recovery" slider, which allows you to bring the highlights back down while keeping the rest of the exposure at normal levels. Or you can do it manually by manipulating the exposure and tone curve. DR is simply a measure of how wide a range of values a camera can capture in a single exposure, and the K-x is capable of capturing a pretty darn wide range. It's up to YOU to know how to process that data to fit that value range into an inherently low-dynamic-range jpeg.

The DR of the K-x and D90 is functionally identical, unless you don't know how to work with the camera's exposure system and optimize exposure in post...

Ecaterin · 03-04-2010, 01:04 PM

I dunno - shooting under real world conditions I haven't seen this problem yet. I've been taking a lot of photos of sunrise clouds the past few weeks - conditions of too bright to look directly at the top of the frame, deep gray clouds at the bottom, every shade of white, gray & gold inbetween. Processing in RAW in Elements I've been able to just slide the exposure recovery bar - not even needing to mess with finer adjustments in curves. Now, if I shoot in RAW+JPG I've seen the K-x clip highlights - my sunrise/clouds photos would not have been salvageable in JPG, but in raw all the data was there with only a little tweaking.

I'm sure in the studio there's a way to make the K-x fail on highlights, but personally I won't be photographing much of anything more difficult for the camera than these sky shots with a huge contrast range. I haven't yet been able to throw *anything* at the K-x that it didn't make look really good - over bright sunrise skies, dark evening snowy mountain against a white/gray sky, racing children, hand held shots of lights in the dark at 1/8....

Wheatfield · 03-04-2010, 04:17 PM

Originally posted by Ecaterin

I'm sure in the studio there's a way to make the K-x fail on highlights, but personally I won't be photographing much of anything more difficult for the camera than these sky shots with a huge contrast range. I haven't yet been able to throw *anything* at the K-x that it didn't make look really good - over bright sunrise skies, dark evening snowy mountain against a white/gray sky, racing children, hand held shots of lights in the dark at 1/8....

You are far less likely to have a DR failure in the studio than in the field, providing you have a whit of knowledge about lighting.
I rarely have more thana 4 stop range in the studio, frequently fun into 6 or more stops in the field.
The thing with people running the camera out of DR (and they get PF this way as well) is to photograph something that is essentially unphotographable.
I always get a laugh out of the fugly pictures of tree branches against a blown out sky to prove that a camera doesn't have sufficient DR, or that a lens has a lot of fringing.
Why not just take normal pictures that you might want to look at rather than creating situations that any camera/lens will fail at and then complain that the equipment failed?

Ecaterin · 03-04-2010, 06:11 PM

Originally posted by Wheatfield

You are far less likely to have a DR failure in the studio than in the field, providing you have a whit of knowledge about lighting.
I rarely have more thana 4 stop range in the studio, frequently fun into 6 or more stops in the field.
The thing with people running the camera out of DR (and they get PF this way as well) is to photograph something that is essentially unphotographable.
I always get a laugh out of the fugly pictures of tree branches against a blown out sky to prove that a camera doesn't have sufficient DR, or that a lens has a lot of fringing.
Why not just take normal pictures that you might want to look at rather than creating situations that any camera/lens will fail at and then complain that the equipment failed?

Wheatfield - sorry, I wasn't clear. For "studio" read "controlled/forced" conditions. I was trying to make the same point - you can make any equipment look bad if you subject it to conditions that are so extreme you will only seldom find them in daily use. Just like I read all the stats about lenses, but I'm much more interested in looking at hundreds of photos by a lens I'm interested in - much more informative to see real usage than charts & graphs of extreme tests.

Class A · 03-04-2010, 06:24 PM

Originally posted by trailrider

For beginners who primarily shoot in Auto mode the tendency to clip highlights can be slightly problematic. For everyone else it's controllable via some manual intervention."

That's all you need to know.
For years DPreview and others have told Pentax off for their conservative metering, leading to, technically, underexposure. Now that the K-x behaves more like other cameras (like the Nikon D40 which needs constant negative exposure compensation or otherwise clips highlights all the time) it is not right either.

In a nutshell, the K-x offers fantastic IQ and DR and it is up to you to expose the way you like it.

As er1kksen has politely pointed out, there is little value to what jlaubza has contributed to the subject.

Wheatfield · 03-04-2010, 06:50 PM

Originally posted by Ecaterin

Wheatfield - sorry, I wasn't clear. For "studio" read "controlled/forced" conditions. I was trying to make the same point - you can make any equipment look bad if you subject it to conditions that are so extreme you will only seldom find them in daily use. Just like I read all the stats about lenses, but I'm much more interested in looking at hundreds of photos by a lens I'm interested in - much more informative to see real usage than charts & graphs of extreme tests.

If we paid attention to the tests, I suspect we would take up basket weaving instead of photography, except that I am sure the rattan would all be defective too.
I see you have a DA35 on the way. I am interested in what you think of it. I have a feeling it might be on my short list.

phzy2003 · 04-07-2010, 10:46 PM

yes, kx is very easy to highlight clips, it is got very good dark but not highlight

Class A · 04-07-2010, 11:31 PM

Originally posted by phzy2003

yes, kx is very easy to highlight clips, it is got very good dark but not highlight

Excuse me, but this doesn't make sense.
In what way would other cameras be better at capturing highlights?

The K-x metering may be less conservative than that of previous Pentax DSLRs but that doesn't mean the camera has a problem with highlights.

The K-x has achieved excellent test results, so if you can add to that, please let us know how to interpret your statement.

jct us101 · 04-08-2010, 12:10 AM

Oh I wouldn't worry too much about the DPreview tests. In real life you probably will never notice this, but DPreview in their infinite wisdom can find a way to make any camera the most awesome that they've ever tested, or the worst of the bunch with just a few modifications to their test scenes. It's quite ingenious really.

03-03-2010, 06:34 PM	#1
trailrider Junior Member Join Date: Mar 2010 Posts: 42	K-X Highlight Clipping? I am new to the forum and considering a K-X purchase. I have read the review at dpreview. I am especially interested in the dynamic range test. Pentax K-x Review: 17. Photographic tests (DR): Digital Photography Review Do people feel the following comment is accurate? If so can it be overcome via adjustments in pre and post processing? "The K-x has a fairly pronounced tendency to clip highlights and over exposed parts of the image are usually not recoverable at all or produce false color information when negative digital exposure compensation is applied in a RAW converter." I am no expert in photography. I have been studying the dynamic range curves. I see just the K-X being just 1/2 EV lower than say a canon. Even when they use their highlight correction. I wouldn't think that would make that much difference. Is there a way to shift the dynamic range curve to the right? To me it looks like the K-X does a better job of shadow recovery. One of the reasons for me in moving to a DSLR is less highlight clipping. Thanks, Mike

03-04-2010, 11:39 AM	#6
jlaubza Senior Member Join Date: Jul 2008 Location: south africa, johannesburg Photos: Gallery Posts: 228	clipping highlights Hi, I tested some RAW images with the K-X in Adobe CS4. Highlights were very clipped and in order to recover these, I had to lower the exposure value of the images quite a lot, ending up with dark images. The limited DR of the KX became very obvious. So, if you get this camera, be prepared for clipped highlights - the DR is visibly less than a Canon or Nikon but the noise factor is better than either the 500d or D90 at the same high iso. At lower iso's (up to 1600) they are much the same. The problem with the K-X is that it has been obviously crippled by cutting out features that would have not made much difference to production cost.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
K5 highlight headroom?	kenyee	Pentax K-5 & K-5 II	10	11-07-2010 01:54 PM
Pentax K-x single channel clipping	gaeriel	Pentax DSLR Discussion	8	09-20-2010 02:27 AM
K-x highlight correction	LN108	Pentax DSLR Discussion	6	07-17-2010 01:48 AM
Clipping on CS3	netuser	Digital Processing, Software, and Printing	4	12-23-2009 08:24 AM
Dynamic Range Extension Shadow Clipping in bokeh	Katsura	Pentax DSLR Discussion	6	12-21-2009 12:15 AM

03-03-2010, 06:48 PM	#2
er1kksen Veteran Member Join Date: Oct 2008 Location: Forestville, NY Photos: Albums Posts: 1,801	Basically, if the DR is the same, regardless of whether there's more in the highlights or shadows, then the DR is the same. The only difference, really, is where on the curve the camera exposes. YOU have control over this: if you're using a Pentax with smaller highlight range and more in the shadows, expose to preserve the highlights and bring the exposure up to whatever's visually "correct" in post. If you're using a Canon with more range in the highlights and less in the shadows, expose to the right (brighter) to take advantage of that range and then bring the exposure down to whatever's visually "correct" in post. The end result is essentially the same.

03-03-2010, 06:51 PM	#3
Wheatfield Moderator Join Date: Apr 2008 Location: The wheatfields of Canada Posts: 15,981	Remember that these tests tend to be designed to test to failure rather than testing to reality. You may find that you have to dial in some exposure comp in certain scenes, but it likely won't matter what brand of camera you are using in this regard..

03-03-2010, 08:37 PM	#4
trailrider Junior Member Join Date: Mar 2010 Posts: 42 Original Poster	I appreciate your replys. I did not study the conclusion information thoroughly. It states: "One of the K-x's main weaknesses is its habit to clip highlights. The problem is exacerbated by a tendency to overexpose high contrast scenes. The K-x also offers very little in terms of RAW headroom. Therefore shooting RAW will not solve the problem entirely. However, the ability to adjust the tone curve during file conversion helps to mitigate this issue. Your best bet though is to apply some negative exposure compensation when necessary. " " All in all the K-x is perfectly capable of producing high quality image output, especially in low light. For beginners who primarily shoot in Auto mode the tendency to clip highlights can be slightly problematic. For everyone else it's controllable via some manual intervention." Mike

03-04-2010, 05:32 AM	#5
rawr Veteran Member Join Date: Jun 2009 Posts: 11,913	It's not something I have noticed terribly much in my use of the camera, tbh. If you shoot in RAW you do actually get a lot of headroom with the K-x. I also think the reviewer needs to think through the fact that every RAW convertor brings it's own 'personality' to the conversion process. Whatever he tried, ACR 2.6 and the included Pentax DCU 4 handles the K-x's RAW output pretty well, highlights and all. Even Picasa and others that use the dcraw engine do OK now that they know about the K-x. While I am not a massive pixel peeper, I still do care about IQ, and I don't see a big issue here.

03-04-2010, 11:57 AM	#7
er1kksen Veteran Member Join Date: Oct 2008 Location: Forestville, NY Photos: Albums Posts: 1,801	jlaubza, the very fact that you were able to recover those highlights demonstrates that the K-x does NOT have limited DR (compared to the 500D and D90). What you need is a program like ACR or lightroom with a "recovery" slider, which allows you to bring the highlights back down while keeping the rest of the exposure at normal levels. Or you can do it manually by manipulating the exposure and tone curve. DR is simply a measure of how wide a range of values a camera can capture in a single exposure, and the K-x is capable of capturing a pretty darn wide range. It's up to YOU to know how to process that data to fit that value range into an inherently low-dynamic-range jpeg. The DR of the K-x and D90 is functionally identical, unless you don't know how to work with the camera's exposure system and optimize exposure in post...

03-04-2010, 01:04 PM	#8
Ecaterin Veteran Member Join Date: Nov 2009 Location: Seattle Photos: Gallery \| Albums Posts: 723	I dunno - shooting under real world conditions I haven't seen this problem yet. I've been taking a lot of photos of sunrise clouds the past few weeks - conditions of too bright to look directly at the top of the frame, deep gray clouds at the bottom, every shade of white, gray & gold inbetween. Processing in RAW in Elements I've been able to just slide the exposure recovery bar - not even needing to mess with finer adjustments in curves. Now, if I shoot in RAW+JPG I've seen the K-x clip highlights - my sunrise/clouds photos would not have been salvageable in JPG, but in raw all the data was there with only a little tweaking. I'm sure in the studio there's a way to make the K-x fail on highlights, but personally I won't be photographing much of anything more difficult for the camera than these sky shots with a huge contrast range. I haven't yet been able to throw anything at the K-x that it didn't make look really good - over bright sunrise skies, dark evening snowy mountain against a white/gray sky, racing children, hand held shots of lights in the dark at 1/8....

03-04-2010, 04:17 PM	#9
Wheatfield Moderator Join Date: Apr 2008 Location: The wheatfields of Canada Posts: 15,981	Originally posted by Ecaterin I'm sure in the studio there's a way to make the K-x fail on highlights, but personally I won't be photographing much of anything more difficult for the camera than these sky shots with a huge contrast range. I haven't yet been able to throw anything at the K-x that it didn't make look really good - over bright sunrise skies, dark evening snowy mountain against a white/gray sky, racing children, hand held shots of lights in the dark at 1/8.... You are far less likely to have a DR failure in the studio than in the field, providing you have a whit of knowledge about lighting. I rarely have more thana 4 stop range in the studio, frequently fun into 6 or more stops in the field. The thing with people running the camera out of DR (and they get PF this way as well) is to photograph something that is essentially unphotographable. I always get a laugh out of the fugly pictures of tree branches against a blown out sky to prove that a camera doesn't have sufficient DR, or that a lens has a lot of fringing. Why not just take normal pictures that you might want to look at rather than creating situations that any camera/lens will fail at and then complain that the equipment failed?

03-04-2010, 06:11 PM	#10
Ecaterin Veteran Member Join Date: Nov 2009 Location: Seattle Photos: Gallery \| Albums Posts: 723	Originally posted by Wheatfield You are far less likely to have a DR failure in the studio than in the field, providing you have a whit of knowledge about lighting. I rarely have more thana 4 stop range in the studio, frequently fun into 6 or more stops in the field. The thing with people running the camera out of DR (and they get PF this way as well) is to photograph something that is essentially unphotographable. I always get a laugh out of the fugly pictures of tree branches against a blown out sky to prove that a camera doesn't have sufficient DR, or that a lens has a lot of fringing. Why not just take normal pictures that you might want to look at rather than creating situations that any camera/lens will fail at and then complain that the equipment failed? Wheatfield - sorry, I wasn't clear. For "studio" read "controlled/forced" conditions. I was trying to make the same point - you can make any equipment look bad if you subject it to conditions that are so extreme you will only seldom find them in daily use. Just like I read all the stats about lenses, but I'm much more interested in looking at hundreds of photos by a lens I'm interested in - much more informative to see real usage than charts & graphs of extreme tests.

03-04-2010, 06:24 PM	#11
Class A Pentaxian Join Date: Aug 2008 Location: Wellington, New Zealand Posts: 11,251	Originally posted by trailrider For beginners who primarily shoot in Auto mode the tendency to clip highlights can be slightly problematic. For everyone else it's controllable via some manual intervention." That's all you need to know. For years DPreview and others have told Pentax off for their conservative metering, leading to, technically, underexposure. Now that the K-x behaves more like other cameras (like the Nikon D40 which needs constant negative exposure compensation or otherwise clips highlights all the time) it is not right either. In a nutshell, the K-x offers fantastic IQ and DR and it is up to you to expose the way you like it. As er1kksen has politely pointed out, there is little value to what jlaubza has contributed to the subject.

03-04-2010, 06:50 PM	#12
Wheatfield Moderator Join Date: Apr 2008 Location: The wheatfields of Canada Posts: 15,981	Originally posted by Ecaterin Wheatfield - sorry, I wasn't clear. For "studio" read "controlled/forced" conditions. I was trying to make the same point - you can make any equipment look bad if you subject it to conditions that are so extreme you will only seldom find them in daily use. Just like I read all the stats about lenses, but I'm much more interested in looking at hundreds of photos by a lens I'm interested in - much more informative to see real usage than charts & graphs of extreme tests. If we paid attention to the tests, I suspect we would take up basket weaving instead of photography, except that I am sure the rattan would all be defective too. I see you have a DA35 on the way. I am interested in what you think of it. I have a feeling it might be on my short list.

04-07-2010, 10:46 PM	#13
phzy2003 Senior Member Join Date: Mar 2010 Location: Gold Coast, Australia Photos: Gallery Posts: 167	yes, kx is very easy to highlight clips, it is got very good dark but not highlight

04-07-2010, 11:31 PM	#14
Class A Pentaxian Join Date: Aug 2008 Location: Wellington, New Zealand Posts: 11,251	Originally posted by phzy2003 yes, kx is very easy to highlight clips, it is got very good dark but not highlight Excuse me, but this doesn't make sense. In what way would other cameras be better at capturing highlights? The K-x metering may be less conservative than that of previous Pentax DSLRs but that doesn't mean the camera has a problem with highlights. The K-x has achieved excellent test results, so if you can add to that, please let us know how to interpret your statement.

04-08-2010, 12:10 AM	#15
jct us101 Veteran Member Join Date: Jun 2008 Location: Rohnert Park, CA Posts: 3,793	Oh I wouldn't worry too much about the DPreview tests. In real life you probably will never notice this, but DPreview in their infinite wisdom can find a way to make any camera the most awesome that they've ever tested, or the worst of the bunch with just a few modifications to their test scenes. It's quite ingenious really.