Better pictures from the ist!

justinr · 03-13-2010, 01:29 PM

I took the K10 out into the field today for the first time and thought I'd got some half decent shots, well I had on the viewing screen but up on the PC they were disappointingly soft.

Some were totally out of focus which I know to be operator error, the main problem being accidental adjustment of the AF point by knocking the button, but the others that should have been good, weren't.

I've just gone back and compared with some similar shots taken with the same lens (50-200 wr) on the ist and found that not only are they sharper overall but also producing larger jpegs. The extra sharpness is partially due to the sharpness setting being set at +1 but that wouldn't account for the smaller jpegs from the bigger camera? Both cameras were set for max jpeg quality.

I'll try and sort some examples out to put up here but I just wonder if anybody has any immediate ideas? I'm flummoxed as well as a little disappointed.

er1kksen · 03-13-2010, 03:10 PM

Could the AF need adjusting?

GoremanX · 03-13-2010, 03:27 PM

What about max JPEG size? Did you have the K10D set to to <whatever its max is>?

Shouldn't really matter, though. If they were both set to 6 MP, I'd expect the same or slightly better quality from the K10D.

justinr · 03-13-2010, 04:19 PM

Looking through the pictures again some more may be explained by not giving time for the camera to focus although I had it switched to tracking mode-

But here there was plenty of time for the AF to work before the shot was taken-

Both images have been cropped and neither is by any means sharp.

The camera was set to record at 10mb, three star jpegs.

GoremanX · 03-13-2010, 04:28 PM

No, they're not sharp, but that's not what jumps out at me. The saturation is very low, I think the white balance was off by a fair bit. If you're using AutoWB, the white horse may be throwing it off. Also, the exposure is way off too. It almost looks like the horse's highlights are blown. This is an oddity, since all that white should've made the picture darker. I can't tell what's going on because the EXIF data got stripped from the pictures.

All of these things can combine to make a picture appear much softer.

Klaus · 03-13-2010, 04:51 PM

Hi, Justin. Fist one looks like there was too much movement (we don't know exposure time). The second one has good focus on the wheels of the car in background, not on the face of the person where you wanted the focus. Which AF-point was selected?
Best regards - Klaus

Marc Sabatella · 03-13-2010, 09:38 PM

In the firts shot, you're dealing with fast motion. Without knowing more about what settings you used and how you took the shot, I'd say it would be crazy to have expected a sharp shot on that. So forget that one.

The second one is kind of small to judge much. Can you post a crop of the area where the focus was? And the identical crop from the same picture picture taken with the same setting with the *ist? If you don't have the identical shot from the *ist, then you need to do a controlled test to get one before coming to any conclusions. As it is, I see no reaosn to assume you're seeing anything other than failure to coose the focus point precisely and/or blowing up your K10D twice as big as your *ist image - *of course* that's going to not look as sharp.

justinr · 03-14-2010, 02:55 AM

Originally posted by Marc Sabatella

In the firts shot, you're dealing with fast motion. Without knowing more about what settings you used and how you took the shot, I'd say it would be crazy to have expected a sharp shot on that. So forget that one.

The second one is kind of small to judge much. Can you post a crop of the area where the focus was? And the identical crop from the same picture picture taken with the same setting with the *ist? If you don't have the identical shot from the *ist, then you need to do a controlled test to get one before coming to any conclusions. As it is, I see no reaosn to assume you're seeing anything other than failure to coose the focus point precisely and/or blowing up your K10D twice as big as your *ist image - *of course* that's going to not look as sharp.

I started off on some control tests last night but it was getting late so not too much was learnt other than the k20 still appeared softer than the ist and was producing a smaller jpeg on exactly the same subject, settings and focal length. The K20 was also producing a brighter image but sensors do vary so no great shakes there.

After sleeping on it my thoughts are that there may be two problems here, the first is an AF gremlin and the second being a firmware issue which may be helped by an update.

All the shots from the whole day varied in sharpness whether action shots or otherwise. By dismissing the first shot shown above is it being suggested that the camera is not suitable for sports work? This is important to me because it will be one of its main roles.

As for the exposure then I tend not to trust AE too much but go by the histogram to find manual settings for a particular light and then adjust as I judge the light to be changing or focal length to be varying. I don't always get it right but I'm not usually far wrong. I was shooting mostly at 400 iso and trying to keep the shutter to 250/320 which meant f5 or 5.6 and even higher when the sun struggled to make it's presence felt. The histogram of the second shot does indicate that things were on the bright side but nothing untoward. Don't be fooled by the pale looking grass, we have had over 70 frosts this winter and very little rain, the countryside really is a weird colour at the moment.

Well I'm off to do some more tests with both bodies and lenses now. I hope I can get this sorted as I really enjoy using the camera.

Justin.

p.s. I was focussing on the horse for the second shot, just foward of the saddle if I remember rightly.

justinr · 03-14-2010, 04:09 AM

Originally posted by GoremanX

No, they're not sharp, but that's not what jumps out at me. The saturation is very low, I think the white balance was off by a fair bit. If you're using AutoWB, the white horse may be throwing it off. Also, the exposure is way off too. It almost looks like the horse's highlights are blown. This is an oddity, since all that white should've made the picture darker. I can't tell what's going on because the EXIF data got stripped from the pictures.

All of these things can combine to make a picture appear much softer.

The saturation, sharpness and contrast are at their default settings and I would agree that they look a little washed out on the monitor but looked quite fine on the LCD (this was the first time I was using it at an event so it was a bit of a shakedown cruise). As I've mentioned (sorry, I forgot that two folk made two different sets of comments when answering Marc) I've never been a fan of AE on any digital camera I've had so I tend to go manual using the histogram as the meter, I tend to get less incorrect exposures that way and I can set for dark or light subjects without trying to confuse myself or the camera by messing about with the exposure compensation or bracketing.

Here is where the method works-

And here is where I got it wrong, the excuse being that I didn't know the colour of the horse until it actually appeared in the viewfinder.

**Rondec** · 03-14-2010, 04:51 AM

I would try doing some focus tests. K10's have a tendency to back focus over time. Pretty easy to fix with the debug mode, but you have to make sure it is the camera and not a particular lens.

Marc Sabatella · 03-14-2010, 08:40 AM

Are you saying that set to 10MP, the K10D file sizes are actually smaller for the same subject? That doesn't seem right at all, so I'd say, forget worrying about focus or any other issues and figure out what menu setting is causing the files to be smaller. No way shol a 10MP JPEG file be smaller than a 6MP one unless you've got a very low quality setting on the former.

justinr · 03-14-2010, 09:24 AM

Originally posted by Marc Sabatella

Are you saying that set to 10MP, the K10D file sizes are actually smaller for the same subject? That doesn't seem right at all, so I'd say, forget worrying about focus or any other issues and figure out what menu setting is causing the files to be smaller. No way shol a 10MP JPEG file be smaller than a 6MP one unless you've got a very low quality setting on the former.

Now this is the other issue I have.

When I put a 1gb card in the slot and set everything to 10mp top quality jpegs then the camera tells me I have room for 207 pictures or appox 4.5 to 5mb per image which is what I'd expect. However, the jpegs I have from it are around 2.5 to 3mb which indicates that it is set on the 6mp setting when most assuredly it is not according to the menu. The jpegs from the K20 tend to be slightly smaller than those from the 6mp ist.

Would a firmware update sort this problem or is it a Pentax repair job do you think?

justinr · 03-14-2010, 09:37 AM

Just as a matter of interest these people were the 'official' event photographers -

www.hoofprints.iel

Click on the All Ireland Hunter Trials 13/3/10 galleries to view their images of the day.

They use Canon 1 or 5D's, not sure which.

GoremanX · 03-14-2010, 01:21 PM

Originally posted by Marc Sabatella

Are you saying that set to 10MP, the K10D file sizes are actually smaller for the same subject? That doesn't seem right at all, so I'd say, forget worrying about focus or any other issues and figure out what menu setting is causing the files to be smaller. No way shol a 10MP JPEG file be smaller than a 6MP one unless you've got a very low quality setting on the former.

A 10MP JPG can be smaller than a 6MP JPG under some circumstances. For example, if the 10MP picture has a bunch of blown highlights or a lot of uniform colours, while the 6MP picture has lots of detail and is very busy. But that's about the only time I can see that being the case. However, I think that may be the case here since so much of these pictures are dominated by a white horse.

Originally posted by justinr

The saturation, sharpness and contrast are at their default settings and I would agree that they look a little washed out on the monitor but looked quite fine on the LCD (this was the first time I was using it at an event so it was a bit of a shakedown cruise).

That may be true, but the fact is, the picture of the horse just standing there clearly has some blown highlights. I'm not talking about the sky, but the horse itself. Regardless of what the histogram said, the exposure there was wrong and detail was lost. It would be interesting to see what other pictures you've taken with this camera that don't have a white horse in them.

Also, I just noticed your filenames start with an underscore (which implies you have the camera set to AdobeRGB), but the posted pictures load up as sRGB. Was there a colour space conversion involved? Or did you re-assign the colour space intentionally? Or did whatever software you use strip out the colour space information and replace it with its own?

It really sucks that there's no EXIF data, that would answer a lot of questions.

edit: the colour space thing might explain all the blown highlights that weren't blown when looking at the histogram.

aleonx3 · 03-14-2010, 01:36 PM

Justinr, I must admit that I am not impressed with those photos - they are either out of focus, over-exposed or look like films with fading color due to age. Also, on the other hand, I am also not impressed with the official photographs taken by "so called" professionals.

There appears to be many different things going wrong that it is tough to get a handle on it without EXIF data.

03-13-2010, 01:29 PM	#1
justinr Veteran Member Join Date: Dec 2009 Location: Tipperary Photos: Gallery Posts: 394	Better pictures from the ist! I took the K10 out into the field today for the first time and thought I'd got some half decent shots, well I had on the viewing screen but up on the PC they were disappointingly soft. Some were totally out of focus which I know to be operator error, the main problem being accidental adjustment of the AF point by knocking the button, but the others that should have been good, weren't. I've just gone back and compared with some similar shots taken with the same lens (50-200 wr) on the ist and found that not only are they sharper overall but also producing larger jpegs. The extra sharpness is partially due to the sharpness setting being set at +1 but that wouldn't account for the smaller jpegs from the bigger camera? Both cameras were set for max jpeg quality. I'll try and sort some examples out to put up here but I just wonder if anybody has any immediate ideas? I'm flummoxed as well as a little disappointed.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Can I use smc Pentax-A lenses on my ist D/ist DS?	Ole	Pentax DSLR and Camera Articles	3	05-18-2007 10:23 PM

03-13-2010, 03:10 PM	#2
er1kksen Veteran Member Join Date: Oct 2008 Location: Forestville, NY Photos: Albums Posts: 1,801	Could the AF need adjusting?

03-13-2010, 03:27 PM	#3
GoremanX Veteran Member Join Date: Dec 2009 Location: Georgia, VT Photos: Albums Posts: 1,657	What about max JPEG size? Did you have the K10D set to to <whatever its max is>? Shouldn't really matter, though. If they were both set to 6 MP, I'd expect the same or slightly better quality from the K10D.

03-13-2010, 04:19 PM	#4
justinr Veteran Member Join Date: Dec 2009 Location: Tipperary Photos: Gallery Posts: 394 Original Poster	Looking through the pictures again some more may be explained by not giving time for the camera to focus although I had it switched to tracking mode- But here there was plenty of time for the AF to work before the shot was taken- Both images have been cropped and neither is by any means sharp. The camera was set to record at 10mb, three star jpegs.

03-13-2010, 04:28 PM	#5
GoremanX Veteran Member Join Date: Dec 2009 Location: Georgia, VT Photos: Albums Posts: 1,657	No, they're not sharp, but that's not what jumps out at me. The saturation is very low, I think the white balance was off by a fair bit. If you're using AutoWB, the white horse may be throwing it off. Also, the exposure is way off too. It almost looks like the horse's highlights are blown. This is an oddity, since all that white should've made the picture darker. I can't tell what's going on because the EXIF data got stripped from the pictures. All of these things can combine to make a picture appear much softer.

03-13-2010, 04:51 PM	#6
Klaus Senior Member Join Date: Jan 2010 Location: Murcia, Spain Photos: Gallery Posts: 100	Hi, Justin. Fist one looks like there was too much movement (we don't know exposure time). The second one has good focus on the wheels of the car in background, not on the face of the person where you wanted the focus. Which AF-point was selected? Best regards - Klaus

03-13-2010, 09:38 PM	#7
Marc Sabatella Veteran Member Join Date: Sep 2006 Location: Denver, CO Photos: Gallery Posts: 10,685	In the firts shot, you're dealing with fast motion. Without knowing more about what settings you used and how you took the shot, I'd say it would be crazy to have expected a sharp shot on that. So forget that one. The second one is kind of small to judge much. Can you post a crop of the area where the focus was? And the identical crop from the same picture picture taken with the same setting with the ist? If you don't have the identical shot from the ist, then you need to do a controlled test to get one before coming to any conclusions. As it is, I see no reaosn to assume you're seeing anything other than failure to coose the focus point precisely and/or blowing up your K10D twice as big as your ist image - of course* that's going to not look as sharp.

03-14-2010, 02:55 AM	#8
justinr Veteran Member Join Date: Dec 2009 Location: Tipperary Photos: Gallery Posts: 394 Original Poster	Originally posted by Marc Sabatella In the firts shot, you're dealing with fast motion. Without knowing more about what settings you used and how you took the shot, I'd say it would be crazy to have expected a sharp shot on that. So forget that one. The second one is kind of small to judge much. Can you post a crop of the area where the focus was? And the identical crop from the same picture picture taken with the same setting with the ist? If you don't have the identical shot from the ist, then you need to do a controlled test to get one before coming to any conclusions. As it is, I see no reaosn to assume you're seeing anything other than failure to coose the focus point precisely and/or blowing up your K10D twice as big as your ist image - of course* that's going to not look as sharp. I started off on some control tests last night but it was getting late so not too much was learnt other than the k20 still appeared softer than the ist and was producing a smaller jpeg on exactly the same subject, settings and focal length. The K20 was also producing a brighter image but sensors do vary so no great shakes there. After sleeping on it my thoughts are that there may be two problems here, the first is an AF gremlin and the second being a firmware issue which may be helped by an update. All the shots from the whole day varied in sharpness whether action shots or otherwise. By dismissing the first shot shown above is it being suggested that the camera is not suitable for sports work? This is important to me because it will be one of its main roles. As for the exposure then I tend not to trust AE too much but go by the histogram to find manual settings for a particular light and then adjust as I judge the light to be changing or focal length to be varying. I don't always get it right but I'm not usually far wrong. I was shooting mostly at 400 iso and trying to keep the shutter to 250/320 which meant f5 or 5.6 and even higher when the sun struggled to make it's presence felt. The histogram of the second shot does indicate that things were on the bright side but nothing untoward. Don't be fooled by the pale looking grass, we have had over 70 frosts this winter and very little rain, the countryside really is a weird colour at the moment. Well I'm off to do some more tests with both bodies and lenses now. I hope I can get this sorted as I really enjoy using the camera. Justin. p.s. I was focussing on the horse for the second shot, just foward of the saddle if I remember rightly.

03-14-2010, 04:09 AM	#9
justinr Veteran Member Join Date: Dec 2009 Location: Tipperary Photos: Gallery Posts: 394 Original Poster	Originally posted by GoremanX No, they're not sharp, but that's not what jumps out at me. The saturation is very low, I think the white balance was off by a fair bit. If you're using AutoWB, the white horse may be throwing it off. Also, the exposure is way off too. It almost looks like the horse's highlights are blown. This is an oddity, since all that white should've made the picture darker. I can't tell what's going on because the EXIF data got stripped from the pictures. All of these things can combine to make a picture appear much softer. The saturation, sharpness and contrast are at their default settings and I would agree that they look a little washed out on the monitor but looked quite fine on the LCD (this was the first time I was using it at an event so it was a bit of a shakedown cruise). As I've mentioned (sorry, I forgot that two folk made two different sets of comments when answering Marc) I've never been a fan of AE on any digital camera I've had so I tend to go manual using the histogram as the meter, I tend to get less incorrect exposures that way and I can set for dark or light subjects without trying to confuse myself or the camera by messing about with the exposure compensation or bracketing. Here is where the method works- And here is where I got it wrong, the excuse being that I didn't know the colour of the horse until it actually appeared in the viewfinder.

03-14-2010, 04:51 AM	#10
Rondec Loyal Site Supporter Join Date: Mar 2009 Location: Gladys, Virginia Photos: Gallery Posts: 27,653	I would try doing some focus tests. K10's have a tendency to back focus over time. Pretty easy to fix with the debug mode, but you have to make sure it is the camera and not a particular lens.

03-14-2010, 08:40 AM	#11
Marc Sabatella Veteran Member Join Date: Sep 2006 Location: Denver, CO Photos: Gallery Posts: 10,685	Are you saying that set to 10MP, the K10D file sizes are actually smaller for the same subject? That doesn't seem right at all, so I'd say, forget worrying about focus or any other issues and figure out what menu setting is causing the files to be smaller. No way shol a 10MP JPEG file be smaller than a 6MP one unless you've got a very low quality setting on the former.

03-14-2010, 09:24 AM	#12
justinr Veteran Member Join Date: Dec 2009 Location: Tipperary Photos: Gallery Posts: 394 Original Poster	Originally posted by Marc Sabatella Are you saying that set to 10MP, the K10D file sizes are actually smaller for the same subject? That doesn't seem right at all, so I'd say, forget worrying about focus or any other issues and figure out what menu setting is causing the files to be smaller. No way shol a 10MP JPEG file be smaller than a 6MP one unless you've got a very low quality setting on the former. Now this is the other issue I have. When I put a 1gb card in the slot and set everything to 10mp top quality jpegs then the camera tells me I have room for 207 pictures or appox 4.5 to 5mb per image which is what I'd expect. However, the jpegs I have from it are around 2.5 to 3mb which indicates that it is set on the 6mp setting when most assuredly it is not according to the menu. The jpegs from the K20 tend to be slightly smaller than those from the 6mp ist. Would a firmware update sort this problem or is it a Pentax repair job do you think?

03-14-2010, 09:37 AM	#13
justinr Veteran Member Join Date: Dec 2009 Location: Tipperary Photos: Gallery Posts: 394 Original Poster	Just as a matter of interest these people were the 'official' event photographers - www.hoofprints.iel Click on the All Ireland Hunter Trials 13/3/10 galleries to view their images of the day. They use Canon 1 or 5D's, not sure which.

03-14-2010, 01:21 PM	#14
GoremanX Veteran Member Join Date: Dec 2009 Location: Georgia, VT Photos: Albums Posts: 1,657	Originally posted by Marc Sabatella Are you saying that set to 10MP, the K10D file sizes are actually smaller for the same subject? That doesn't seem right at all, so I'd say, forget worrying about focus or any other issues and figure out what menu setting is causing the files to be smaller. No way shol a 10MP JPEG file be smaller than a 6MP one unless you've got a very low quality setting on the former. A 10MP JPG can be smaller than a 6MP JPG under some circumstances. For example, if the 10MP picture has a bunch of blown highlights or a lot of uniform colours, while the 6MP picture has lots of detail and is very busy. But that's about the only time I can see that being the case. However, I think that may be the case here since so much of these pictures are dominated by a white horse. Originally posted by justinr The saturation, sharpness and contrast are at their default settings and I would agree that they look a little washed out on the monitor but looked quite fine on the LCD (this was the first time I was using it at an event so it was a bit of a shakedown cruise). That may be true, but the fact is, the picture of the horse just standing there clearly has some blown highlights. I'm not talking about the sky, but the horse itself. Regardless of what the histogram said, the exposure there was wrong and detail was lost. It would be interesting to see what other pictures you've taken with this camera that don't have a white horse in them. Also, I just noticed your filenames start with an underscore (which implies you have the camera set to AdobeRGB), but the posted pictures load up as sRGB. Was there a colour space conversion involved? Or did you re-assign the colour space intentionally? Or did whatever software you use strip out the colour space information and replace it with its own? It really sucks that there's no EXIF data, that would answer a lot of questions. edit: the colour space thing might explain all the blown highlights that weren't blown when looking at the histogram.

03-14-2010, 01:36 PM	#15
aleonx3 Veteran Member Join Date: Jul 2007 Location: Brampton, Ontario Photos: Gallery \| Albums Posts: 3,996	Justinr, I must admit that I am not impressed with those photos - they are either out of focus, over-exposed or look like films with fading color due to age. Also, on the other hand, I am also not impressed with the official photographs taken by "so called" professionals. There appears to be many different things going wrong that it is tough to get a handle on it without EXIF data.