Where Am I Screwing Up? (3 Images)

marlon · 10-12-2008, 05:24 AM

Hi folks,

I spent 5 hours in blazing heat yesterday, just to come home once again, and find that about 4 of the 400 photos I took are usable.

Im not sure why this happens, and Im not sure if it's camera shake or OOF. Maybe you could have a look at these three 100% crops (no editing done) and help me?

Note that I used spot AF on the main subject in each of these.

Im finding it hard to believe that it is camera shake, because as you will see, in most cases my shutter speed was many times more than the FL/SS rule expects it to be.

On the other hand, I trusted AF.s mode the whole day, but AF never failed me before, so why would it this time?

I used two sigma lenses all day (EX DC 18-50 and 70-300 APO) and both showed the same results.

One last thing to note, I used the ISO D-range feature with ISO set to 200. Could this be it? What does this exactly do?

Hope to hear from someone soon

Appreciated, regards
Marlon

1/2000 f/4.0 50mm ISO 200 (D-range)

1/500 f/4.0 18mm ISO 200 (D-range)

1/1000 f/6.7 100mm ISO 400 (not D-range)

jmc7104 · 10-12-2008, 05:50 AM

I am not familuar with the K20 but that is not camera shake. It almost has to be focus problem but doesnt seem too appear in your photo album. I would check the contact pins between the lenses and camera for dirt or possibly collapsed. to happen to 2 lenses the same day would indicate to me that it was a connection to the camera or a bad drive motor (Focus)

tomtor · 10-12-2008, 06:20 AM

Originally posted by marlon

Hi folks,

I spent 5 hours in blazing heat yesterday, just to come home once again, and find that about 4 of the 400 photos I took are usable.

Im not sure why this happens, and Im not sure if it's camera shake or OOF. Maybe you could have a look at these three 100% crops (no editing done) and help me?

Note that I used spot AF on the main subject in each of these.

Image #1 DOES have fine detail in the bottom center leaf. You can see the small vessels which are about a single pixel in this 100% crop. Also the leave slightly more up/right shows detail, so the center focus point could have picked one of these two leaves. The focus area is bigger than you expect.

The same applies to the center bottom and central leaf in #3, you can see texture. It is not unfocussed.

#2 is clearly unfocussed, no explanation for that. Is it a center crop or near the edge? Can you show a resized total image?

The images also don't have much contrast, did you use a hood on your lenses?

marlon · 10-12-2008, 06:39 AM

tomtor & jmc thanks.

Tom, it seems what you say makes sense. I might be overly critical, and should have used manual focus on difficult images like #1.

I did indeed use a hood, but must confess I have a seriously cheap Kenko UV filter on my lenses (same type on both). I also realized that with the opened up to 4.0 this might also have affected the sharpness.
I agree that the contrast is low, after I read your explanation it started to make sense.

I also searched the forums now on the center AF point being bigger than I'd expect. Didn't know that.
As for #2, ive never had much luck focussing on leafy stuff like trees, very hard with the viewfinder being so small

Thanks again for the input

Wheatfield · 10-12-2008, 06:58 AM

Just a wild guess, but it looks to me like your lens just isn't sharp at f/4, and that the filter is making things worse.
The picture you shot at f/6.7 is almost, kind of sharp, but thats the best that can be said.
You might also have focusing problems, but I am presuming you did have things in focus when you pushed the button.

tomtor · 10-12-2008, 07:20 AM

Originally posted by marlon

tomtor & jmc thanks.

Tom, it seems what you say makes sense. I might be overly critical, and should have used manual focus on difficult images like #1.

I did indeed use a hood, but must confess I have a seriously cheap Kenko UV filter on my lenses (same type on both). I also realized that with the opened up to 4.0 this might also have affected the sharpness.
I agree that the contrast is low, after I read your explanation it started to make sense.

I also searched the forums now on the center AF point being bigger than I'd expect. Didn't know that.
As for #2, ive never had much luck focussing on leafy stuff like trees, very hard with the viewfinder being so small

Thanks again for the input

Can't find the diagram now (did a search too), but all of the focus areas are so big that they almost touch each other. Some even partially overlap if I remember it correctly. The dot is just the approximate location of the center of each area. You can easily test this by focusing on a small object on a white piece of paper. Even when the object is quite far from a focus dot, it snaps into focus. Try this in automatic focus point selection mode.

This means that sometimes when you have a small subject, your camera will focus on an object in the background.

Manual focus does not really help in my experience without a dedicated focus screen with a split. I find that if my camera focuses on the right subject, it IS often far more accurate than my manual focus attempts, at least with fast lenses (F2.8 or faster). So I just try to verify if the camera did pick the right subject by inspecting the viewfinder carefully, but in general I let the AF-unit do the focusing.

Marc Sabatella · 10-12-2008, 12:02 PM

I would agree that for the most part, these look like pretty much normal results for 100% crops when shooting wide open with relatively inexpensive lenses. The sharp areas area as sharp as one could reaosnably expect for 100% crops taken under such conditions, but between the AF point not being *precisely* where you wanted (and this is not something you can really control), and DOF not being as deep as you might have wanted (this much you could have controleld by stopping down further), the overall result is indeed the focus being a bit off.

Aside from stopping fown more and/or buying more expensive lenses, only other thing I'd suggest is getting the O-ME53 magnifying eyecup if you are having trouble focusing manually (and if that doens't do the trick, invest in a Katz Eye split focus screen).

The only exception here is #2, which does look more completely out of focus. I'd need to see the whole image to understand better what might have happened there. On the other hand, maybe the lens is just really soft at 18mm.

Peter Zack · 10-12-2008, 01:12 PM

I agree with Marc that at 100% you are seeing the limits of these lenses. Adding a cheap UV will only compound the issue. Then a shallow DOF and you have a softer looking picture. Remember that each lens has a 'best aperture' area and that is generally the middle range of f8 to f11. Plus zoom lenses tend to be worst at the extremes. So you've added all the worst aspects together and gotten what appears to be a soft image. I would want to see the full images to be certain there is any issue.

At ISO 200 the camera will be fine and produce sharp images. The D range is an attempt to preserve highlight detail and reduce blown out areas of the shot. It works well but is subtle. It should have little or no effect on focus.

If you really want to be sure that the system is working, get a tripod. Lock it down firmly. Use the 2 second MLU (or cable release with SR off) and choose a subject that is static, less monotone or less 'busy'. AF the camera on the subject and take the test shots. The image should be about as sharp as the lens can render. If the normal sized shots appear soft then, you might have an issue but I suspect not.

Last avoid assuming 100% crops of the image will look awesome at this magnification. Even the best lenses will look a little soft at this level. You will need to get used to what a sharply focused shot looks like at 100%. Good but not great. Once you get familiar with what to look for and expect, use the 100% to check focus only and go back to normal size.

Ole · 10-12-2008, 01:41 PM

Your best bet is to stop down to around f/8. If the shutter speed gets too low, bump the ISO.

You may also want to experiment with Shake Reduction on/off. I'm getting less than stellar results with the K20D and SR on, if the shutterspeed is somewhat fast.

axl · 10-12-2008, 02:57 PM

I agree with what is written above. But I have a question:
in no.2 what were you focusing on?

mithrandir · 10-12-2008, 05:28 PM

I notice a small branch in the lower left hand quadrant that seems more in focus. I wonder if it fooled the AF.

NicholasN · 10-12-2008, 06:20 PM

marlon,

My first K20D body exhibited the same issues, I returned the camera and tests revealed the AF sensors were no good and they replaced the camera.

I would send it back if I were you.

tomtor · 10-12-2008, 11:12 PM

Originally posted by mithrandir

I notice a small branch in the lower left hand quadrant that seems more in focus. I wonder if it fooled the AF.

I think that is indeed the cause!

Well spotted!

marlon · 10-13-2008, 12:17 AM

Hi

Thanks to all who replied, Peter Zack, Marc, and those on the same line of thought appears to be correct. It seems like my AF was fooled indeed, I didn't know how big the af sensing areas were, I had the lens at the extremes and it was a damn hot day, so there was alot of discomfort adding to the experience (making wierd angle poses harder than it should be).

I played around last night, doing AF and MF ons various objects and cars speeding past etc, and had no issues.

So between the above issues and my inexperience, I found the problem. Im too scared to use more ISO, but will learn to do so.

I feel much better now, thanks guys.

PS: axl, I focussed with the middle af area on one of the conifer trees(where it meets the sky). Seems this Sigma EX is pretty soft at that aperture and 18mm

srudolph · 10-18-2008, 06:08 AM

Marlon,

Pic #2, I think you may have focused on the tree, but the camera focused on the sky. With the lens at 18mm, you were close to the trees, at f4 even a 18mm lense will have shallow depth of field. Best course in this situation would be to drop shutter speed to about 1/50-1/60, and close down the f-stop.

When you have time, play with the DOF preview on the shutter release. Learn how the lense sees. I know this is harder with zooms than with primes, but it is a study worth while.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
K-7 iso 3200 images: No NR, yet the faces look plastic. (11 images)	pcarfan	Pentax DSLR Discussion	43	08-31-2010 08:13 AM
Pentax utility screwing with my pics???	timstone	Digital Processing, Software, and Printing	2	05-30-2010 02:45 PM
A variety of images with FA* 200/4 Macro (no bugs) - VLF competition images	Marc Langille	Post Your Photos!	28	08-22-2008 07:28 PM
High ISO concert images with Tam 28-300 (Images)	jsundin	Post Your Photos!	2	07-05-2007 08:19 AM
A few K10D tourist images of Italy (9 images, about 200kb each)	volosong	Post Your Photos!	2	04-24-2007 04:29 AM

10-12-2008, 05:24 AM	#1
marlon Inactive Account Join Date: Aug 2008 Location: Cape Town, South Africa Posts: 232	Where Am I Screwing Up? (3 Images) Hi folks, I spent 5 hours in blazing heat yesterday, just to come home once again, and find that about 4 of the 400 photos I took are usable. Im not sure why this happens, and Im not sure if it's camera shake or OOF. Maybe you could have a look at these three 100% crops (no editing done) and help me? Note that I used spot AF on the main subject in each of these. Im finding it hard to believe that it is camera shake, because as you will see, in most cases my shutter speed was many times more than the FL/SS rule expects it to be. On the other hand, I trusted AF.s mode the whole day, but AF never failed me before, so why would it this time? I used two sigma lenses all day (EX DC 18-50 and 70-300 APO) and both showed the same results. One last thing to note, I used the ISO D-range feature with ISO set to 200. Could this be it? What does this exactly do? Hope to hear from someone soon Appreciated, regards Marlon 1/2000 f/4.0 50mm ISO 200 (D-range) 1/500 f/4.0 18mm ISO 200 (D-range) 1/1000 f/6.7 100mm ISO 400 (not D-range)

10-12-2008, 05:50 AM	#2
jmc7104 Veteran Member Join Date: Feb 2008 Location: Rock Island, Illinois 61201 Photos: Gallery Posts: 436	I am not familuar with the K20 but that is not camera shake. It almost has to be focus problem but doesnt seem too appear in your photo album. I would check the contact pins between the lenses and camera for dirt or possibly collapsed. to happen to 2 lenses the same day would indicate to me that it was a connection to the camera or a bad drive motor (Focus)

10-12-2008, 06:20 AM	#3
tomtor Veteran Member Join Date: Apr 2008 Location: Netherlands Posts: 382	Originally posted by marlon Hi folks, I spent 5 hours in blazing heat yesterday, just to come home once again, and find that about 4 of the 400 photos I took are usable. Im not sure why this happens, and Im not sure if it's camera shake or OOF. Maybe you could have a look at these three 100% crops (no editing done) and help me? Note that I used spot AF on the main subject in each of these. Image #1 DOES have fine detail in the bottom center leaf. You can see the small vessels which are about a single pixel in this 100% crop. Also the leave slightly more up/right shows detail, so the center focus point could have picked one of these two leaves. The focus area is bigger than you expect. The same applies to the center bottom and central leaf in #3, you can see texture. It is not unfocussed. #2 is clearly unfocussed, no explanation for that. Is it a center crop or near the edge? Can you show a resized total image? The images also don't have much contrast, did you use a hood on your lenses?

10-12-2008, 06:39 AM	#4
marlon Inactive Account Join Date: Aug 2008 Location: Cape Town, South Africa Posts: 232 Original Poster	tomtor & jmc thanks. Tom, it seems what you say makes sense. I might be overly critical, and should have used manual focus on difficult images like #1. I did indeed use a hood, but must confess I have a seriously cheap Kenko UV filter on my lenses (same type on both). I also realized that with the opened up to 4.0 this might also have affected the sharpness. I agree that the contrast is low, after I read your explanation it started to make sense. I also searched the forums now on the center AF point being bigger than I'd expect. Didn't know that. As for #2, ive never had much luck focussing on leafy stuff like trees, very hard with the viewfinder being so small Thanks again for the input

10-12-2008, 06:58 AM	#5
Wheatfield Moderator Join Date: Apr 2008 Location: The wheatfields of Canada Posts: 15,981	Just a wild guess, but it looks to me like your lens just isn't sharp at f/4, and that the filter is making things worse. The picture you shot at f/6.7 is almost, kind of sharp, but thats the best that can be said. You might also have focusing problems, but I am presuming you did have things in focus when you pushed the button.

10-12-2008, 07:20 AM	#6
tomtor Veteran Member Join Date: Apr 2008 Location: Netherlands Posts: 382	Originally posted by marlon tomtor & jmc thanks. Tom, it seems what you say makes sense. I might be overly critical, and should have used manual focus on difficult images like #1. I did indeed use a hood, but must confess I have a seriously cheap Kenko UV filter on my lenses (same type on both). I also realized that with the opened up to 4.0 this might also have affected the sharpness. I agree that the contrast is low, after I read your explanation it started to make sense. I also searched the forums now on the center AF point being bigger than I'd expect. Didn't know that. As for #2, ive never had much luck focussing on leafy stuff like trees, very hard with the viewfinder being so small Thanks again for the input Can't find the diagram now (did a search too), but all of the focus areas are so big that they almost touch each other. Some even partially overlap if I remember it correctly. The dot is just the approximate location of the center of each area. You can easily test this by focusing on a small object on a white piece of paper. Even when the object is quite far from a focus dot, it snaps into focus. Try this in automatic focus point selection mode. This means that sometimes when you have a small subject, your camera will focus on an object in the background. Manual focus does not really help in my experience without a dedicated focus screen with a split. I find that if my camera focuses on the right subject, it IS often far more accurate than my manual focus attempts, at least with fast lenses (F2.8 or faster). So I just try to verify if the camera did pick the right subject by inspecting the viewfinder carefully, but in general I let the AF-unit do the focusing.

10-12-2008, 12:02 PM	#7
Marc Sabatella Veteran Member Join Date: Sep 2006 Location: Denver, CO Photos: Gallery Posts: 10,685	I would agree that for the most part, these look like pretty much normal results for 100% crops when shooting wide open with relatively inexpensive lenses. The sharp areas area as sharp as one could reaosnably expect for 100% crops taken under such conditions, but between the AF point not being precisely where you wanted (and this is not something you can really control), and DOF not being as deep as you might have wanted (this much you could have controleld by stopping down further), the overall result is indeed the focus being a bit off. Aside from stopping fown more and/or buying more expensive lenses, only other thing I'd suggest is getting the O-ME53 magnifying eyecup if you are having trouble focusing manually (and if that doens't do the trick, invest in a Katz Eye split focus screen). The only exception here is #2, which does look more completely out of focus. I'd need to see the whole image to understand better what might have happened there. On the other hand, maybe the lens is just really soft at 18mm.

10-12-2008, 01:12 PM	#8
Peter Zack Pentaxian Moderator Emeritus Join Date: May 2007 Location: Edmonton Alberta, Canada Photos: Gallery \| Albums Posts: 10,643	I agree with Marc that at 100% you are seeing the limits of these lenses. Adding a cheap UV will only compound the issue. Then a shallow DOF and you have a softer looking picture. Remember that each lens has a 'best aperture' area and that is generally the middle range of f8 to f11. Plus zoom lenses tend to be worst at the extremes. So you've added all the worst aspects together and gotten what appears to be a soft image. I would want to see the full images to be certain there is any issue. At ISO 200 the camera will be fine and produce sharp images. The D range is an attempt to preserve highlight detail and reduce blown out areas of the shot. It works well but is subtle. It should have little or no effect on focus. If you really want to be sure that the system is working, get a tripod. Lock it down firmly. Use the 2 second MLU (or cable release with SR off) and choose a subject that is static, less monotone or less 'busy'. AF the camera on the subject and take the test shots. The image should be about as sharp as the lens can render. If the normal sized shots appear soft then, you might have an issue but I suspect not. Last avoid assuming 100% crops of the image will look awesome at this magnification. Even the best lenses will look a little soft at this level. You will need to get used to what a sharply focused shot looks like at 100%. Good but not great. Once you get familiar with what to look for and expect, use the 100% to check focus only and go back to normal size.

10-12-2008, 01:41 PM	#9
Ole Administrator Join Date: Sep 2006 Location: Copenhagen, Denmark Photos: Gallery \| Albums Posts: 4,408	Your best bet is to stop down to around f/8. If the shutter speed gets too low, bump the ISO. You may also want to experiment with Shake Reduction on/off. I'm getting less than stellar results with the K20D and SR on, if the shutterspeed is somewhat fast.

10-12-2008, 02:57 PM	#10
axl Veteran Member Join Date: Mar 2007 Location: Nove Zamky, Slovakia Photos: Gallery \| Albums Posts: 7,183	I agree with what is written above. But I have a question: in no.2 what were you focusing on?

10-12-2008, 05:28 PM	#11
mithrandir Veteran Member Join Date: Nov 2007 Location: Maryland Photos: Gallery Posts: 1,895	I notice a small branch in the lower left hand quadrant that seems more in focus. I wonder if it fooled the AF.

10-12-2008, 06:20 PM	#12
NicholasN Inactive Account Join Date: May 2008 Location: Sydney, Australia Posts: 301	marlon, My first K20D body exhibited the same issues, I returned the camera and tests revealed the AF sensors were no good and they replaced the camera. I would send it back if I were you.

10-12-2008, 11:12 PM	#13
tomtor Veteran Member Join Date: Apr 2008 Location: Netherlands Posts: 382	Originally posted by mithrandir I notice a small branch in the lower left hand quadrant that seems more in focus. I wonder if it fooled the AF. I think that is indeed the cause! Well spotted!

10-13-2008, 12:17 AM	#14
marlon Inactive Account Join Date: Aug 2008 Location: Cape Town, South Africa Posts: 232 Original Poster	Hi Thanks to all who replied, Peter Zack, Marc, and those on the same line of thought appears to be correct. It seems like my AF was fooled indeed, I didn't know how big the af sensing areas were, I had the lens at the extremes and it was a damn hot day, so there was alot of discomfort adding to the experience (making wierd angle poses harder than it should be). I played around last night, doing AF and MF ons various objects and cars speeding past etc, and had no issues. So between the above issues and my inexperience, I found the problem. Im too scared to use more ISO, but will learn to do so. I feel much better now, thanks guys. PS: axl, I focussed with the middle af area on one of the conifer trees(where it meets the sky). Seems this Sigma EX is pretty soft at that aperture and 18mm

10-18-2008, 06:08 AM	#15
srudolph New Member Join Date: Sep 2006 Location: Western New York, USA Posts: 9	Marlon, Pic #2, I think you may have focused on the tree, but the camera focused on the sky. With the lens at 18mm, you were close to the trees, at f4 even a 18mm lense will have shallow depth of field. Best course in this situation would be to drop shutter speed to about 1/50-1/60, and close down the f-stop. When you have time, play with the DOF preview on the shutter release. Learn how the lense sees. I know this is harder with zooms than with primes, but it is a study worth while.