Focus Issue

I posted this on FM, but didn't receive a response. Maybe I'm not seeing clearly. Please read and if you have a moment, view and comment. Thanks, in advance.

Hi,

If anyone has just a few moments, please go to this site (see link below) and have a look. I am worried I may have a bad copy of a lens, or MAYBE (hopefully) it is me. I am having difficulty getting a nice crisp shot with this lens. I purchased it here on buy and sell, so it is not new. It is the Canon 135 f/2.8L USM on the Canon 5D body.

While I realize some of the lighting wasn't picture perfect, I had little choice, first of all and second of all, that is less the issue for me than the focus. I am reading the 5D user manual again to make sure I have my setting correctly programmed.

AF mode was AI Focus
I used Evaluative Metering

Regarding the focus points, if I select, for example, the far right point, is it true that ALL features in the same distance as where that point hits SHOULD BE IN FOCUS AS WELL, EVEN IF IT IS ON THE FAR LEFT OF THE VIEW FINDER?

http://Di.smugmug.com/gallery/2808717/1/150006154

Thanks so much for your input.

Not really sure I see a focus issue. Judging from the fact that the subject fills the frame you must have been fairly close with a 135mm lens. At f4 I'm guessing that the depth of field is pretty darn small. At 100% I can see areas on several of these that are definitely sharp, but it may not be the point you want. If the subject is not flat relative to you then parts of it will be less sharp as they get further from the focus point, as would be the case when looking up at the subject in a tree?

If you open these with the Canon supplied image viewer, which I admit I don't use, doesn't it show which focus point was used? Personally I find that I tend to change my settings so that only the central focus point is used.

Kevin

One responder mentioned that use of the 5d with the 135 2.0 has a VERY short dof. I may have missed this child's eyes by anywhere from two inches to 12. I think I may have focused on the tree once or twice and he was back from the tree by several inches. Hopefully that is what this is about.

I am accustomed to tremendously sharp focus with the 30D and 24-105 L lens. This was a bit puzzling to me. Now I am really conscious of setting the focus point directly on the subjects eyes which means I sacrifice timing. I may miss that very second that caught my attention and I wanted to grab with the lens because I was am so busy making sure I line up the focus with the eyes.

Maybe with practice.

Thanks very much for your response. I am getting little feedback on this thread and I would love to solve this mystery.

Diana

It might be worth setting up some practice scenes with a subject that extends front and back so you can confirm just where the focal point is? It may be that once you get comfortable with the lens at different apertures you'll feel more in control of the results.

Good luck

Kevin

I have only had a chance to look at one shot, Ezra9.jpg. Unfortunately, Sumgmug's "large" is very small and the "original" only allows me to see part of the image at a time.

Have you tried viewing the image in EOS Viewer? It allows you to see which AF points were used by the camera.

When I look at the focus plane, on the tree on the left, it gives me some idea of the DOF, and it would certainly appear that the eyes are behind the in focus area by a few centimeters.

With the 25-105 mm on the 30D you would have more DOF. This image is also at f/4, which would have been the maximum aperture of the 24-105 mm.

Brian A.

Try using 'one shot' for still subjects. There are too many variables that can come into play that can screw up your photos when using AI focus. When using such a shallow dof I would select the center focus point only. Then use the focus & recompose technique.

--
-Steve
'Only a poor craftsman blames his tools'

http://www.stevewhitlockphotography.com

I looked at your first original of the child in the tree...my opinion is that you are using too high an iso and/or underexposing...there is too much noise in that photo....even 800 iso should give you much less noise. Other posters are correct that the depth of field in the 5d is much shallower than the cropped cameras. For portraits, I would generally stop down to at least F4.
--
Andy C
http://www.echelonphotographers.com

First of all, I did think to do a one shot metering. So I have considered this possibility.

As far as the ISO, it was under 400 for ALL the shots. I tried to emphasize that I am aware of some of the overexposure and that's not what I'm concerned about. I am concerned that the lens is a bit soft. But I still have some testing to do. The tree limb may have moved the focus to several centimeters foward of his eyes. I am hoping this is what's going on...

By the way, all f stops are 4.0 and 5.6 in all of these pictures. 1/500 & 1/1000s as well.

Thank you for the feedback. It has been helpful. I think there are really two issues. One is that this copy of the lens is a fair copy and maybe showing its age and two is the shooter. Ahh yes, that would be me. I need to practice using this lens with the 5D...the dof is extremely shallow.

thanks again.

I don't see a focus issue. A couple of folks say the depth of field is more shallow on the 5D, it is not. It is the same as any other body using the same lens at the same distance. Assuming you are about 10-15 feet away from your subject, and at the apertures you are saying you used, your depth of field will be between 5-20 inches, depending of course on the exact distance from subject and f-stop. If you shot the same shots with a 30D or 1DMKIII, from the same spot with the same settings, your depth of field would not change.

If you are focusing on the tree or the childs face or body, you'll see more of the front of the tree out of focus. About 1/3 of your focus field will be in front of the focus point, the other 2/3 behind it. If your f-stop is F4, and you are focused on the child, you don't have much room to play with here (5 inches or so)

Stop down if you need more subject in focus.
--
Visit me at

http://www.have-camera-will-travel.com/

Hmmm, what you say re the DOF makes a lot of sense. The reason I believed other posters' statements that the DOF is VERY Shallow is because it is the full format sensor. So the size of the sensor, the lack of crop factor lead me to the quick conclusion that this is so. In fact, I am taking my 30D out for a shoot with the 135 lens this afternoon to see if there's a difference. Oh well.

Anyway, I don't necessarily agree with the conclusion there's no problem with the focus on 90% of these shots. Most likely the depth of field on this lens itself is more shallow than I typically work with, and I need to practice my focus if I want to use this lens. OR, I'm wondering if it's in need of a fix...

The focus is off simply by a few inches. If you look at some of these, you can see every crisp detail of the bark on the tree, but the boys eyes are soft. Some of them show the entire face soft. I think I need to set the Automatic Focus Mode onto single shot and that may help freeze my focus for shots like this. If it's a moving object, obviously I need to use the AI Focus mode. But for individual portraits, I should try single shot mode.

Thanks for your help. By the way, there's a formula to calculate DOF distances based on several factors. Are you familiar with this? It's on DOFmaster.com. I tried to download it but I'm having difficulty putting it into my palm.

Diana

Radio Denver said:

A couple of folks say the depth of
field is more shallow on the 5D, it is not. It is the same as any
other body using the same lens at the same distance.

Click to expand...

Were it only true. To get the same framing with the 5D as the 30D, which is the comparison being made, the subject to sensor plane distance would be considerably shorter – and the DOF would also be much shallow.

Brian A.

I've never been able to get every shot in perfect focus no matter how hard I try using auto-focus on the 5D. I always set the focus for the * button using the custom functions, that way I can do a quick auto-focus on the subject, then change the fov for framing and it keeps the camera from refocusing when I set exposure. I'm partial to lenses that allow a manual focus with the auto-focus turned on because of this. Truthfully though, I find that I go manual focus most of the time on shots like this. Old habits are hard to break.
--
Visit me at

http://www.have-camera-will-travel.com/

It's all a question of distance to subject. If you are standing in a particular position, it doesn't matter what camera body you use. The dof is going to be the same with any lense at that distance. If you move closer to the subject, the depth of field will remain identical with either body. The DOF myth is a reversal of logic in an attempt to equate a full frame body with a crop sensor. The point is, the lenses were designed for 35mm full frame cameras, it's the crop bodies that don't work the same way.

When you're in the field and you only have a single spot to work from, it doesn't matter which camera you use, the calculation and estimate of proper depth of field will remain the same for either body. The crop sensor just chops off the outer third of your image. If you need less field of view, use a longer lens and adjust the aperture to get the correct depth of field you are looking for. Otherwise, your adjusting the rules to fit your misconception and you'll end up getting it wrong more times than not.

Hugowolf said:

Radio Denver said:

A couple of folks say the depth of
field is more shallow on the 5D, it is not. It is the same as any
other body using the same lens at the same distance.

Click to expand...

Were it only true. To get the same framing with the 5D as the 30D,
which is the comparison being made, the subject to sensor plane
distance would be considerably shorter – and the DOF would also be
much shallow.

Brian A.

Click to expand...

--
Visit me at

http://www.have-camera-will-travel.com/

OK, according to an article on the following site:

http://www.Cambridgeincolour.com/tutorials/depth-of-field.html

DOF calculations involve either sensor size or film size. Therefore, the conclusion that the 5D has a different dof calculation from the 30D for example is correct according to this article.

Diana

Perhaps you aren't interpreting what I'm saying correctly or you aren't interpreting what the tutorial is saying correctly.

From a photographiclly practical standpoint, I'll give you the following situation.

You are in a city standing 30 feet in front of a large statue.
You have in your possession two camera bodies.
1 EOS 5D
1 EOS 30D

You have mounted on the EOS 5D a 50mm prime lens.
You have no lens mounted on the EOS 30D.

You frame the statue with the 5D, set your aperture to f8 and snap the shot.

You remove the lens from the 5D and place it on the 30D.

You do not move.

You frame the statue in the 30D at f8 and snap the shot.

You haven't moved.

The DOF is the same in both shots.

If you back up to make the 30D shot fill the frame the same as the 5D frames, then your DOF will change.

Both bodies, with the same lens, from the same position, at the same f-stop will give the same DOF.

Diana Keat said:

OK, according to an article on the following site:

http://www.Cambridgeincolour.com/tutorials/depth-of-field.html

DOF calculations involve either sensor size or film size.
Therefore, the conclusion that the 5D has a different dof
calculation from the 30D for example is correct according to this
article.

Diana

Click to expand...

--
Visit me at

http://www.have-camera-will-travel.com/

You may have a slight AF calibration issue. You can try doing an AF test using this AF test target and scale. Here's the standard setup:
http://www.canon-dslr.com/Canon_Jan05/Canon_SLR_Focus_Test.htm

The highlighted hairs are crisp and sharp so there is no focus issue. Youmigth not like the depth of field, go to higher f-stops.,

The hairs are sharp, but without knowing where she focused on you can't say for sure whether her AF calibration is ok. I think she said that she was focusing on the tree bark to compinsate for a few shots. It's still worth the effort to at least rule out any calibration issues.

wkay said:

The highlighted hairs are crisp and sharp so there is no focus
issue. Youmigth not like the depth of field, go to higher f-stops.,

Click to expand...