What's wrong with my k-5 raw images?

outsider · 04-16-2011, 10:47 AM

I've done a lot of close observation of the K-5's raw images and have noticed something very apparent. The K-5's images (at least the raw, images which is what I shoot) are significantly softer than the K-x's images. I shoot both in aperture priority with as similar settings as could be (with settings that do and do NOT apply to raw files). I like to use the eye area for a measure of what the camera/lens is doing (accurate focus, contrast, sharpness). since that's what I aim for and what i'm used to analyzing closely at the end of the day. But looking at several images from the k-x and k-5, shot under similar conditions and settings, the k-5 is much hazier, less contrasty. Reflections are muted, contrast is more subdued (in every closeup i've looked at). It's so significant, I'm thinking there may be something wrong. $1300. this is NOT what I was expecting. I understand the k-5's sensor is new and renders a certain look, but this was a surprise. Has anyone else noticed this? Are there things i'm not considering that I should be? (increased DR shouldn't do this). I really want to keep this camera. Test images were all shot with the "contrasty" sigma 85mm 1.4. Sorry for the underexposure. The blue eye is the k-x image, the green eye, the k-5.

Attached Images

View Picture EXIF

View Picture EXIF

aurele · 04-16-2011, 11:08 AM

first of all, the focusing is well done on the eye with the k-x not on the k5. that's the first point.
(maybe you have FF or BF ? the k5 allow you to adjust it. look in the menu.)

are you sure that the generals settings are the same ? (on the k-x, info button / personalize image / normal, portrait, landscape, etc .. you can adjust sharpness for exemple.)

HEEGZ · 04-16-2011, 11:09 AM

I actually prefer the K5 image and only see a slight softness in comparison. There is a difference in shutter speed as well. If you are that concerned about your K5 you might do a few test images on a tripod to put your mind at ease. If you want to make a good comparison you should shoot from a tripod and use all of the same settings. I don't think you can make an accurate comparison with the two images you have here, and even so, the K5 image looks better to me. My guess is that this is just a slight movement on your part, or your subject moved, or maybe that your lens was wide open. I can't imagine that at a normal viewing resolution or size that any of this would be noticeable. I'm curious now if a static object shot from a tripod at identical settings would produce the difference you seem to notice.

outsider · 04-16-2011, 11:22 AM

Originally posted by aurele

first of all, the focusing is well done on the eye with the k-x not on the k5. that's the first point.
(maybe you have FF or BF ? the k5 allow you to adjust it. look in the menu.)

are you sure that the generals settings are the same ? (on the k-x, info button / personalize image / normal, portrait, landscape, etc .. you can adjust sharpness for exemple.)

Dude (not you, just excited), the k-5 image was 1 of about 3 or 4, close as possible, recomposing the shot as well as humanly possible. It is the best example i'm capable of. If you told me to go and do it again, we may see the same results. Did some more tests last night on high contrast image, focus seems ok. Also, none of the settings you mention affect raw. There is information that goes into the raw data (raw "remembers" the cameras jpeg settings) that can transfer those settings to jpeg, but ONLY if you use the camera to convert (which I don't) or you use pentax computer software (which I don't). I use adobe photoshop CS4. I don't have enough reason yet to mess with the lens adjustment, never have so I don't feel comfortable yet. But it is hard to tell if the image is just hazy, milky.....or out of focus.

Last edited by outsider; 04-16-2011 at 11:34 AM.

outsider · 04-16-2011, 11:31 AM

Originally posted by HEEGZ

I actually prefer the K5 image and only see a slight softness in comparison. There is a difference in shutter speed as well. If you are that concerned about your K5 you might do a few test images on a tripod to put your mind at ease. If you want to make a good comparison you should shoot from a tripod and use all of the same settings. I don't think you can make an accurate comparison with the two images you have here, and even so, the K5 image looks better to me. My guess is that this is just a slight movement on your part, or your subject moved, or maybe that your lens was wide open. I can't imagine that at a normal viewing resolution or size that any of this would be noticeable. I'm curious now if a static object shot from a tripod at identical settings would produce the difference you seem to notice.

Thanks for the reply. The k-5 image was shot at 1/640, the k-5, about 1000 (the exif info should be accessible by tab at top of picture). Probably wouldnt be noticeable without blowing up, but I am a bit obsessive

Hmmm. Well, i don't know. I can only assume that most people don't pixel peep like I do. I do need to remember that the k-5 DOES purple fringe more than normal (so much so, that people are widely complaining), so this "haziness" could be another "side effect" of this new, powerful sensor. Time will tell I spose.

outsider · 04-16-2011, 11:40 AM

If someone has the K-5 and K-x, maybe they could take a picture in similar scenarios (raw only), maybe something they're used to seeing. Be nice to find out if this is me, or other owners are so excited about their k-5, they just haven't noticed it.

HEEGZ · 04-16-2011, 11:48 AM

I've not kept up with all of the issues people have had with the K5, but I would think PF is more of a lens issue than a body problem. Also, you should be able to fix that in CS4 if it is bad enough people complain. It looks like your K-x image was at 1/3200, but I'm not sure how relevant that is. You have a lot of close face shots on your Flickr page so I can see why this is important to you. I'd be interested in seeing some full shots you've taken with the K5 that are not satisfactory, because everything I could find looks like it was taken with your K-x.

On a side note, I try not to pixel peep at all. I end up finding flaws that I never notice when I'm looking at the full image, and generally just discourage myself about either my gear or my technique. I suppose it is important if there is a problem with your K5, but I can't see one based off of a single cropped image. BTW, do you have another lens you could use for comparison?

qtopplings · 04-16-2011, 11:59 AM

just return it, or stop with these ridiculous threads. No way in hell the K-x beats the K-5 in anything, except for price.

DSims · 04-16-2011, 12:25 PM

Your results agree exactly with my experience in going from the K-x to the K-5. I did get one of the early production K-5s (but I don't believe that makes much difference). I still haven't decided for sure which I like better, but I thought I wanted my K-x back at first (I sold it just before getting the K-5). So you're not imagining things - this is representative of the actual differences between the cameras.

Honestly, I think my solution has been that I pixel-peep less now. I have seen situations in which the 16 Mp is nice, and it does appear (as people say) that the K-5 can recover information well from RAW when I have exposed a scene wrong.

I still haven't decided for sure which I like better IQ-wise. Some would say the K-5 images look more professional. In some cases I think it's because they actually prefer the images, but in others I wonder if it's just because the K-5 is "supposed" to be more professional.

In any case, it's taken me a while to warm up to my K-5, but I think I'm appreciating it more now. I'm not yet sure if I'm "maturing" in my tastes or simply getting used to it.

bc_the_path · 04-16-2011, 12:30 PM

Your K5 is front focusing.
Just look at the eyebrows; they are sharp in your second photo and blurry in your first...
Must be very easy to adjust.

DSims · 04-16-2011, 12:30 PM

A number of posters have said the image looks out of focus, and that's just the point - typical results from the K-5 can "look" out of focus when viewed close up, even if tremendous effort is put into getting correct focus. This look seems to appear even when the K-5 is "properly" focused. On my K-5 the focusing screen is much less accurate than LV, I've noticed, but LV is not practical in many applications. I have a horrible time getting good focus at or near wide open with my K50/1.2 on my K-5, but not on my K200D. So it may be an actual FF issue, but it's hard to overcome. This is the main reason I'm thinking of sending my K-5 in for repair - to have the focusing screen adjusted.

I'd like to see more feedback on this issue from others who have owned both the K-5 and the K-x or K-r.

outsider · 04-16-2011, 12:34 PM

Originally posted by DSims

Your results agree exactly with my experience in going from the K-x to the K-5. I did get one of the early production K-5s (but I don't believe that makes much difference). I still haven't decided for sure which I like better, but I thought I wanted my K-x back at first (I sold it just before getting the K-5). So you're not imagining things - this is representative of the actual differences between the cameras.

Honestly, I think my solution has been that I pixel-peep less now. I have seen situations in which the 16 Mp is nice, and it does appear (as people say) that the K-5 can recover information well from RAW when I have exposed a scene wrong.

I still haven't decided for sure which I like better IQ-wise. Some would say the K-5 images look more professional. In some cases I think it's because they actually prefer the images, but in others I wonder if it's just because the K-5 is "supposed" to be more professional.

In any case, it's taken me a while to warm up to my K-5, but I think I'm appreciating it more now. I'm not yet sure if I'm "maturing" in my tastes or simply getting used to it.

Fantastic! Thanks for you input. So this IS the K-5 look. You are not alone. My serial number starts with 402XXXX!!! So it seems they all are like this. Good to know it's probably not a flaw. Yes, it will take some getting used to. The k-x is just a sweet camera. I'm almost thinking about getting another one since my shutter count is at just over 27,000 (in about 10 months). But I really appreciate your input. I.Q. wise, i'm not convinced the k-5 is much better. I've had it only about 3 days, and i've learned a lot, but this is such a sophisticated machine, time will give me more opportunity to learn and appreciate it. I do noticed that images look better.....hard to explain......more dynamic, deeper, richer look to them when they are viewed as a final product, that is, from afar (not pixel peeping). Hope to hear more on peoples opinion on this "look". Thanks again

outsider · 04-16-2011, 12:38 PM

Originally posted by qtopplings

just return it, or stop with these ridiculous threads. No way in hell the K-x beats the K-5 in anything, except for price.

Haha Man

Right on! I'm not stopping sh*t until I find out what's going on with this k-5. My k-5 lens combination is well over $2000. This forum is free. Pixel peepers, get in here!

outsider · 04-16-2011, 12:45 PM

Here's another crop of the focus point, the eye again. Exif info should be here. Shot from about 5 feet back. Much softer raw results than i've gotten used to in the k-x. It is focusing accurately. This is much softer point of focus at 1.4, with the k-5 than 1.4 using the k-x with the sigma 85 on both cameras. There would be much more sheen on the eye and contrast in the hair and skin otherwise. I wonder if pentax will ever address this in an update. Or maybe this type of rendering is better for post processing?

bc_the_path · 04-16-2011, 12:56 PM

I do not understand why you don't see the point.
If you are interested in the "look" of K5 images; shoot an eye at f/8 and examine (pixel-peep) the image. If you are not interested in whether the shallow DOF (combined with some FF) is effecting the sharpness of your images; why are you shooting wide open?

04-16-2011, 12:30 PM	#11
DSims Veteran Member Join Date: Sep 2009 Location: Southern California Photos: Gallery \| Albums Posts: 3,236	A number of posters have said the image looks out of focus, and that's just the point - typical results from the K-5 can "look" out of focus when viewed close up, even if tremendous effort is put into getting correct focus. This look seems to appear even when the K-5 is "properly" focused. On my K-5 the focusing screen is much less accurate than LV, I've noticed, but LV is not practical in many applications. I have a horrible time getting good focus at or near wide open with my K50/1.2 on my K-5, but not on my K200D. So it may be an actual FF issue, but it's hard to overcome. This is the main reason I'm thinking of sending my K-5 in for repair - to have the focusing screen adjusted. I'd like to see more feedback on this issue from others who have owned both the K-5 and the K-x or K-r. Last edited by DSims; 04-16-2011 at 12:41 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
damaged files? what is wrong with my images?	pete_pf	Pentax DSLR Discussion	63	10-27-2011 01:14 PM
RAW DNG Shots with Wrong White Balance Settings	nstocke	Troubleshooting and Beginner Help	8	09-10-2010 02:36 AM
upload raw images	WTurnerphotography	Troubleshooting and Beginner Help	4	03-19-2010 11:25 AM
RAW duplicate images - saving as RAW files	jpzk	Digital Processing, Software, and Printing	17	02-15-2010 07:01 PM
DxO doesn't correct images/ Am I doing sth. wrong?	Egg Salad	Digital Processing, Software, and Printing	9	01-28-2010 05:35 PM

04-16-2011, 11:08 AM	#2
aurele Veteran Member Join Date: Jan 2011 Location: Paris, France Photos: Albums Posts: 3,217	first of all, the focusing is well done on the eye with the k-x not on the k5. that's the first point. (maybe you have FF or BF ? the k5 allow you to adjust it. look in the menu.) are you sure that the generals settings are the same ? (on the k-x, info button / personalize image / normal, portrait, landscape, etc .. you can adjust sharpness for exemple.)

04-16-2011, 11:09 AM	#3
HEEGZ Senior Member Join Date: Jun 2010 Location: Missouri Photos: Albums Posts: 258	I actually prefer the K5 image and only see a slight softness in comparison. There is a difference in shutter speed as well. If you are that concerned about your K5 you might do a few test images on a tripod to put your mind at ease. If you want to make a good comparison you should shoot from a tripod and use all of the same settings. I don't think you can make an accurate comparison with the two images you have here, and even so, the K5 image looks better to me. My guess is that this is just a slight movement on your part, or your subject moved, or maybe that your lens was wide open. I can't imagine that at a normal viewing resolution or size that any of this would be noticeable. I'm curious now if a static object shot from a tripod at identical settings would produce the difference you seem to notice.

04-16-2011, 11:22 AM	#4
outsider Veteran Member Join Date: Jan 2011 Photos: Albums Posts: 416 Original Poster	Originally posted by aurele first of all, the focusing is well done on the eye with the k-x not on the k5. that's the first point. (maybe you have FF or BF ? the k5 allow you to adjust it. look in the menu.) are you sure that the generals settings are the same ? (on the k-x, info button / personalize image / normal, portrait, landscape, etc .. you can adjust sharpness for exemple.) Dude (not you, just excited), the k-5 image was 1 of about 3 or 4, close as possible, recomposing the shot as well as humanly possible. It is the best example i'm capable of. If you told me to go and do it again, we may see the same results. Did some more tests last night on high contrast image, focus seems ok. Also, none of the settings you mention affect raw. There is information that goes into the raw data (raw "remembers" the cameras jpeg settings) that can transfer those settings to jpeg, but ONLY if you use the camera to convert (which I don't) or you use pentax computer software (which I don't). I use adobe photoshop CS4. I don't have enough reason yet to mess with the lens adjustment, never have so I don't feel comfortable yet. But it is hard to tell if the image is just hazy, milky.....or out of focus. Last edited by outsider; 04-16-2011 at 11:34 AM.

04-16-2011, 11:31 AM	#5
outsider Veteran Member Join Date: Jan 2011 Photos: Albums Posts: 416 Original Poster	Originally posted by HEEGZ I actually prefer the K5 image and only see a slight softness in comparison. There is a difference in shutter speed as well. If you are that concerned about your K5 you might do a few test images on a tripod to put your mind at ease. If you want to make a good comparison you should shoot from a tripod and use all of the same settings. I don't think you can make an accurate comparison with the two images you have here, and even so, the K5 image looks better to me. My guess is that this is just a slight movement on your part, or your subject moved, or maybe that your lens was wide open. I can't imagine that at a normal viewing resolution or size that any of this would be noticeable. I'm curious now if a static object shot from a tripod at identical settings would produce the difference you seem to notice. Thanks for the reply. The k-5 image was shot at 1/640, the k-5, about 1000 (the exif info should be accessible by tab at top of picture). Probably wouldnt be noticeable without blowing up, but I am a bit obsessive Hmmm. Well, i don't know. I can only assume that most people don't pixel peep like I do. I do need to remember that the k-5 DOES purple fringe more than normal (so much so, that people are widely complaining), so this "haziness" could be another "side effect" of this new, powerful sensor. Time will tell I spose.

04-16-2011, 11:40 AM	#6
outsider Veteran Member Join Date: Jan 2011 Photos: Albums Posts: 416 Original Poster	If someone has the K-5 and K-x, maybe they could take a picture in similar scenarios (raw only), maybe something they're used to seeing. Be nice to find out if this is me, or other owners are so excited about their k-5, they just haven't noticed it.

04-16-2011, 11:48 AM	#7
HEEGZ Senior Member Join Date: Jun 2010 Location: Missouri Photos: Albums Posts: 258	I've not kept up with all of the issues people have had with the K5, but I would think PF is more of a lens issue than a body problem. Also, you should be able to fix that in CS4 if it is bad enough people complain. It looks like your K-x image was at 1/3200, but I'm not sure how relevant that is. You have a lot of close face shots on your Flickr page so I can see why this is important to you. I'd be interested in seeing some full shots you've taken with the K5 that are not satisfactory, because everything I could find looks like it was taken with your K-x. On a side note, I try not to pixel peep at all. I end up finding flaws that I never notice when I'm looking at the full image, and generally just discourage myself about either my gear or my technique. I suppose it is important if there is a problem with your K5, but I can't see one based off of a single cropped image. BTW, do you have another lens you could use for comparison?

04-16-2011, 11:59 AM	#8
qtopplings Senior Member Join Date: Nov 2009 Location: Marin, CA Photos: Gallery \| Albums Posts: 282	just return it, or stop with these ridiculous threads. No way in hell the K-x beats the K-5 in anything, except for price.

04-16-2011, 12:25 PM	#9
DSims Veteran Member Join Date: Sep 2009 Location: Southern California Photos: Gallery \| Albums Posts: 3,236	Your results agree exactly with my experience in going from the K-x to the K-5. I did get one of the early production K-5s (but I don't believe that makes much difference). I still haven't decided for sure which I like better, but I thought I wanted my K-x back at first (I sold it just before getting the K-5). So you're not imagining things - this is representative of the actual differences between the cameras. Honestly, I think my solution has been that I pixel-peep less now. I have seen situations in which the 16 Mp is nice, and it does appear (as people say) that the K-5 can recover information well from RAW when I have exposed a scene wrong. I still haven't decided for sure which I like better IQ-wise. Some would say the K-5 images look more professional. In some cases I think it's because they actually prefer the images, but in others I wonder if it's just because the K-5 is "supposed" to be more professional. In any case, it's taken me a while to warm up to my K-5, but I think I'm appreciating it more now. I'm not yet sure if I'm "maturing" in my tastes or simply getting used to it.

04-16-2011, 12:30 PM	#10
bc_the_path Veteran Member Join Date: Mar 2007 Location: Ankara, Turkey Photos: Gallery Posts: 401	Your K5 is front focusing. Just look at the eyebrows; they are sharp in your second photo and blurry in your first... Must be very easy to adjust.

04-16-2011, 12:34 PM	#12
outsider Veteran Member Join Date: Jan 2011 Photos: Albums Posts: 416 Original Poster	Originally posted by DSims Your results agree exactly with my experience in going from the K-x to the K-5. I did get one of the early production K-5s (but I don't believe that makes much difference). I still haven't decided for sure which I like better, but I thought I wanted my K-x back at first (I sold it just before getting the K-5). So you're not imagining things - this is representative of the actual differences between the cameras. Honestly, I think my solution has been that I pixel-peep less now. I have seen situations in which the 16 Mp is nice, and it does appear (as people say) that the K-5 can recover information well from RAW when I have exposed a scene wrong. I still haven't decided for sure which I like better IQ-wise. Some would say the K-5 images look more professional. In some cases I think it's because they actually prefer the images, but in others I wonder if it's just because the K-5 is "supposed" to be more professional. In any case, it's taken me a while to warm up to my K-5, but I think I'm appreciating it more now. I'm not yet sure if I'm "maturing" in my tastes or simply getting used to it. Fantastic! Thanks for you input. So this IS the K-5 look. You are not alone. My serial number starts with 402XXXX!!! So it seems they all are like this. Good to know it's probably not a flaw. Yes, it will take some getting used to. The k-x is just a sweet camera. I'm almost thinking about getting another one since my shutter count is at just over 27,000 (in about 10 months). But I really appreciate your input. I.Q. wise, i'm not convinced the k-5 is much better. I've had it only about 3 days, and i've learned a lot, but this is such a sophisticated machine, time will give me more opportunity to learn and appreciate it. I do noticed that images look better.....hard to explain......more dynamic, deeper, richer look to them when they are viewed as a final product, that is, from afar (not pixel peeping). Hope to hear more on peoples opinion on this "look". Thanks again

04-16-2011, 12:38 PM	#13
outsider Veteran Member Join Date: Jan 2011 Photos: Albums Posts: 416 Original Poster	Originally posted by qtopplings just return it, or stop with these ridiculous threads. No way in hell the K-x beats the K-5 in anything, except for price. Haha Man Right on! I'm not stopping sh*t until I find out what's going on with this k-5. My k-5 lens combination is well over $2000. This forum is free. Pixel peepers, get in here!

04-16-2011, 12:56 PM	#15
bc_the_path Veteran Member Join Date: Mar 2007 Location: Ankara, Turkey Photos: Gallery Posts: 401	I do not understand why you don't see the point. If you are interested in the "look" of K5 images; shoot an eye at f/8 and examine (pixel-peep) the image. If you are not interested in whether the shallow DOF (combined with some FF) is effecting the sharpness of your images; why are you shooting wide open?