7D - is it the camera or the lens?

Olympus 8mm f/1.8 M.Zuiko Digital ED Fisheye PRO Lens

09/21/2011 01:20:34 PM · #51

I did get focus confirmation. My camera is set not to shoot unless it has found focus. And that shot isn't macro, I'm sure the strip was at least 6 inches high. I actually got focus confirmation several times that would have produced similar results, but only bothered to take two blurred shots and two that were relatively decent.

DCNUTTER

Canon EF-S 15-85mm f/3.5-5.6 IS USM

09/21/2011 03:04:58 PM · #52

The minimum focus distance of that lens is .46 meters. So make sure you're outside that distance no matter what the focal length. If you are physically pretty close to an object and using a mid to long focal length your DOF will be very shallow. It you add to that low contrast subject in low light you are going to have focus problems even with fairly small aperture settings.

Some of your focus points fall on less than ideal contrast levels, and you shouldn't ever focus on the leaves of a tree like that because even the slightest movement of the leaves will give a poor focus result, and leaves are layered and go deep into the tree so you really don't know where it's focusing on. As far as the last shots you posted, I tend to agree with JH in that it looks like you might have been close to the edge of the minimum focus distance along with a shallow DOF.

In regard to focusing cards...In order to accurately use them you have to use a formula to determine the ideal distance from the sensor to the focus card which is dependent of the focal length setting of your lens. Just picking arbitrary apertures, focal lengths, and distance to the card is not the correct way to test the lens. So if you plan on using one multiply your focal length your using by 50 and divide by 1000 to give you the distance in meters you should be from the test chart. So for a prime 50mm lens for instance you would want to be 2.5 meters or just over 8 foot away. For a 200mm focal length you would ideally want to be 10 meters away or 32 feet away for accurate adjustments. You can tell if your front or back focusing, and cameras such as the 7D can be calibrated lens to body using the micro-adjustment settings for focus. I don't think that is your issue here.

How does the lens focus in good light using a middle aperture such as f/11.0, a 50mm focal length with a distance of about 8 to 10 foot from the object....when focusing on a high contrast area, and using a shutter speed fast enough to hand hold....meaning at least 1/100 sec at that focal length.

Dave

Olympus 8mm f/1.8 M.Zuiko Digital ED Fisheye PRO Lens

09/21/2011 03:39:27 PM · #53

Dave I just checked the Metadata - the focal distance of the decent shot is 1.1 metre, so not too close. The bad shot shows the focal distance as 0.4 metre, so that is way out.

It's dark already here, so I think I will have to leave doing more tests until tomorrow. But it does appear that I've ruled out the possibility that this is a lens problem.

Edit for clarity: Although the Metadata shows such big differences in focal distance, the two photos were taken from approximately the same distance, as can be seen from the similarity in framing. Both were taken with the zoom fully extended and this is the full frame, resized, not a crop.

Message edited by author 2011-09-21 19:34:04.

WHBYankee

Canon EOS-500D Rebel T1i

Canon EF 24-70mm f/2.8L USM

09/21/2011 03:45:01 PM · #54

Don't forget to turn off Image Stabilization when you use a tripod. It could cause "shake" or blurred effect.

Olympus 8mm f/1.8 M.Zuiko Digital ED Fisheye PRO Lens

09/21/2011 04:34:46 PM · #55

Originally posted by WHBYankee:

Don't forget to turn off Image Stabilization when you use a tripod. It could cause "shake" or blurred effect.

I mentioned somewhere in the thread that I had done so. I've made that mistake on occasion, but I've always picked up the problem when I've checked my first shot. The latest photos however were taken hand-held with an unstabilised lens.

What worries me about today's shots is not that the camera couldn't focus on my subject (some of the time) but that it thought it could and produced results so far out.

SteveJ

Canon EF-S 18-135mm f/3.5-5.6 IS STM

09/21/2011 05:06:16 PM · #56

I am confused now, your camera is under warranty and has been in for repairs. It still hasn't been fixed, so why haven't you been back to where you bought it and asked for a replacement camera?? Surely if this is an ongoing problem that Canon hasn't fixed, you are entitled to either a proper repair or a replacement.

I know that I wouldn't be going through what you are. If Canon haven't fixed it, I would be onto the shop and demanding a new camera.

Olympus 8mm f/1.8 M.Zuiko Digital ED Fisheye PRO Lens

09/21/2011 05:29:12 PM · #57

SteveJ the repair was for the camera freezing, which was probably only a firmware issue. It was only after I got it back that I started worrying about focus. I've always assumed that focus errors where cause by mistakes on my part until I got it back and suddenly noticed focus problems with photos taken on a tripod with mirror lock-up etc. What I'm trying to establish now is whether in fact I have a focus problem as some people here seem to feel the camera is okay. I don't feel like going back to the service centre and making a fool of myself if there's nothing wrong with the camera.

SteveJ

Canon EF-S 18-135mm f/3.5-5.6 IS STM

09/21/2011 05:34:07 PM · #58

Originally posted by GinaRothfels:

This I can understand, but you have until December to get this sorted out. How are other lenses working? Is it just the 18-55mm that plays up? I would rather look a fool and walk out with a properly working camera than maintain my aplomb with a duff camera.

Olympus 8mm f/1.8 M.Zuiko Digital ED Fisheye PRO Lens

09/21/2011 05:44:56 PM · #59

Originally posted by SteveJ:

No, I think I've cleared the possibility of it being a lens issue. Today I was using the 18-55mm, but when I first noticed the problem over the weekend I was using the Sigma 18-200mm IS. I specially changed lenses to see whether the problem went away, but it didn't. The problem is either the camera or an incompetent user.

As things stand at the moment I will probably take a couple of days to do some more tests and if I still feel the camera is at fault, I will go back to the service centre next week.

09/21/2011 05:58:46 PM · #60

Ok..hold on...

Nutter says:

The minimum focus distance of that lens is .46 meters. So make sure you're outside that distance no matter what the focal length. If you are physically pretty close to an object and using a mid to long focal length your DOF will be very shallow. It you add to that low contrast subject in low light you are going to have focus problems even with fairly small aperture settings.

And you say

Dave I just checked the Metadata - the focal distance of the decent shot is 1.1 metre, so not too close. The bad shot shows the focal distance as 0.4 metre, so that is way out.

Isn't this problem solved at this point?

Olympus 8mm f/1.8 M.Zuiko Digital ED Fisheye PRO Lens

09/21/2011 06:04:25 PM · #61

kenskid it doesn't seem right to me that the camera should think it has found focus so far from the correct distance. It's suggestions that I'm trying to focus on things that are too difficult that makes me wonder if I'm expecting too much of the camera. But with what the 7D cost, I'd like to believe it's better than the results it's producing right now.

ETA: Rereading your comment I'm not sure whether you were saying the bad shot was taken from too close when in fact both were taken from approximately the same distance, as can been seen from the similarity in framing.

Message edited by author 2011-09-21 18:06:35.

09/21/2011 06:10:32 PM · #62

So both shots were with the same lens from the same distance from the object?

Originally posted by GinaRothfels:

Olympus 8mm f/1.8 M.Zuiko Digital ED Fisheye PRO Lens

09/21/2011 06:14:27 PM · #63

Originally posted by kenskid:

So both shots were with the same lens from the same distance from the object?

Yes. There may have been an inch or two difference, though the framing would suggest not. These are not crops, they are the full frame.

09/21/2011 06:18:13 PM · #64

...and the metadata said 1.1 meter for the good one and .40 for the bad?

Originally posted by GinaRothfels:

Originally posted by kenskid:
So both shots were with the same lens from the same distance from the object?

Yes. There may have been an inch or two difference, though the framing would suggest not. These are not crops, they are the full frame.

Olympus 8mm f/1.8 M.Zuiko Digital ED Fisheye PRO Lens

09/21/2011 06:34:02 PM · #65

Originally posted by kenskid:

...and the metadata said 1.1 meter for the good one and .40 for the bad?

Correct.

09/21/2011 08:23:04 PM · #66

Then the metadata shows .40 means that it was too close to the subject if that lens' closest focus distance is .46 right?

Now if you're telling me you set a tripod up and shot one shot and then another...and the meta showed 1.1 for the first and .40 for the second, then there is something major wrong with the sensors/optics of the camera/lens. The difference between 1.1 meter and .40 meters is more than 18 inches. Did you move the camera that much between shots?

Originally posted by GinaRothfels:

Originally posted by kenskid:
...and the metadata said 1.1 meter for the good one and .40 for the bad?

Correct.

Olympus 8mm f/1.8 M.Zuiko Digital ED Fisheye PRO Lens

09/21/2011 08:34:36 PM · #67

Originally posted by kenskid:

I took the bad shot first, then a few minutes later I decided I should try to get a good one as a comparison. Though the camera was not on a tripod, I was still sitting on the same spot and judging by the similarity in framing of the two shots, I'm confident that the real distance was almost the same.

MelonMusketeer

Nikon D7000

Nikon MF 50mm f/1.4

09/21/2011 10:02:33 PM · #68

Your test card is a good example of one of the things that the camera would normally have a tough time getting correct A F lock with.
Look carefully at the A F section in your camera manual. There are a few paragraphs and illustrations about A F problems with repeating vertical or horizontal lines or geometric patterns.
The camera may be "seeing" two different points in the pattern that are similar, as you can do by looking cross eyed at vertical lines, and triangulating the focus solution as if it is seeing one point closer or more distant than it actually is.

DCNUTTER

Canon EF-S 15-85mm f/3.5-5.6 IS USM

09/21/2011 10:41:10 PM · #69

Gina, I'm not sure what that strip is is those photos, but for one thing it's looks reflective. Your eyes might be able to see the contrast of those little diamond shapes, but a sensor trying to determine a distinct contrast difference might be thrown off by too much light being reflected back. Since the photos are not cropped and the strip takes up most of the frame it's safe to assume you were focusing in the center since the focus points wouldn't naturally fall along the extreme edges of the viewfinder/frame.

If you were standing greater than the minimum focus distance away from the strip, you set your camera to single manual focus point (auto focus points are not reliable for this particular shot)and have the focus point sit half on the white or lighter wall area and half on the strip (since it appears a bit darker) your focus point will have more contrast between the two areas to focus with. Since it looks like a relatively flat surface meaning the strip seems pretty flush to the wall or surface it's on an aperture of between f/8 to f/11 should do well. As long as you have a fast enough shutter speed you can hand hold. To be on the safe side make sure your shutter speed is at least 2 times the focal length setting on your lens....or faster.

So if you're shooting at 90mm focal length you want your shutter speed to be 1/200 sec on the Canon camera or faster. If you're set at an aperture of f/8 or f/11, you're the proper distance way and you have the shutter speed take the shot. If you can't get the proper shutter speed bump the ISO up until you can with a good balance between image quality and noise. If you get too much noise by doing that then lower your ISO back down and put the camera on your tripod.

You really need to take what I would consider more traditional shot with good lighting and good contrast along with following all the rest of the guidelines I suggested to really determine if the camera is at fault or if it's just how and what you are shooting. The 7D focus system is actually really great, and personally I haven't found a lot of complaints about it. I certainly don't have any. Since you get some good shots, and some that aren't I really don't think from everything I've seen so far that it's the camera.

I personally don't use the auto focus points. In fact I'm pretty much a manual setting guy. I normally shoot in full manual mode and I will use auto focus on the lens when I'm hand holding, but I still always select the focus point manually. If you are used to using auto focus point selection...I don't care what kind of camera it is...you WILL get a lot more missed focus shots because the camera often focuses on the wrong thing. It doesn't matter if it's a Nikon, Canon, Point and Shoot or high end DSLR. It's the nature of the beast. Cameras are dumb boxes that take their best guess, but they don't know where you want the focus. If a scene has multiple things that fall around the focus points it's hit or miss. With manual focus points I never miss focus with my 7D. Now having the best aperture at that moment for the best DOF is another matter entirely and that of course comes with experience for anyone.

I would take your camera out on a nice sunny day, give yourself some shooting room where you are not on the minimum focus distance edge, fast shutter speeds and focus on a persons nearest eye with at least an aperture of f/5.6 to ensure a decent DOF. Set it up on your tripod and take a shot of a decent sized wall clock..preferably with a nice white clock face and dark hands. Set the focus point on one of the hands which has great contrast against the face and shoot. I just don't think a lot of your subjects in the test shots are ideal if you're trying to determine if your camera focus is an issue.

I know you said earlier that you try to take more test shots today. I'll check back later tomorrow to see what kind of results you got (if you had the time to test it out). Good luck.

Dave

Apple iPhone 5

Nikon AF-S DX Zoom-Nikkor 18-55mm f/3.5-5.6 ED

09/22/2011 06:14:40 AM · #70

Originally posted by DCNUTTER:

If you are used to using auto focus point selection...I don't care what kind of camera it is...you WILL get a lot more missed focus shots because the camera often focuses on the wrong thing.

That's the part that confuses me about these test shots. If my camera can't get focus, it hunts around and no matter how much I press the shutter, it won't fire.

In that test shot, the camera focused on 'something' at 0.40m distance and fired. But what did it focus on? (I can't see anything in sharp focus in the shot)

james_so

Canon EOS 77D

Sigma 30mm f/1.4 EX DC HSM for Canon

09/22/2011 07:24:01 AM · #71

Originally posted by JH:

Originally posted by DCNUTTER:
If you are used to using auto focus point selection...I don't care what kind of camera it is...you WILL get a lot more missed focus shots because the camera often focuses on the wrong thing.

That's the part that confuses me about these test shots. If my camera can't get focus, it hunts around and no matter how much I press the shutter, it won't fire.

In that test shot, the camera focused on 'something' at 0.40m distance and fired. But what did it focus on? (I can't see anything in sharp focus in the shot)

Just because it thinks it has found a focus point doesn't mean it has. This also explains the different focus distances recorded in the EXIF, the camera thought it had a focus at .40 but obviously was confused (probably by the pattern).

Best suggestion is to go out and shoot stuff, take lots of photos in good light and see what the results are, if they aren't satisfactory take it in and have someone look at it or send it back and get a replacement.

Olympus 8mm f/1.8 M.Zuiko Digital ED Fisheye PRO Lens

09/22/2011 08:22:39 AM · #72

Originally posted by james_so:

Originally posted by JH:

Originally posted by DCNUTTER:
If you are used to using auto focus point selection...I don't care what kind of camera it is...you WILL get a lot more missed focus shots because the camera often focuses on the wrong thing.

That's the part that confuses me about these test shots. If my camera can't get focus, it hunts around and no matter how much I press the shutter, it won't fire.

In that test shot, the camera focused on 'something' at 0.40m distance and fired. But what did it focus on? (I can't see anything in sharp focus in the shot)

Just because it thinks it has found a focus point doesn't mean it has. This also explains the different focus distances recorded in the EXIF, the camera thought it had a focus at .40 but obviously was confused (probably by the pattern).

Best suggestion is to go out and shoot stuff, take lots of photos in good light and see what the results are, if they aren't satisfactory take it in and have someone look at it or send it back and get a replacement.

I don't use auto focus point selection, occasionally I'll use zone selection, but usually I choose my own focal point. In this case everything was at the same distance so any focal point should have got the same result.

JH the reason you can't see anything in focus is because there was NOTHING 0.40m away from the camera. You've expressed the exact concern I had - if the camera can't find focus it normally won't let me take the shot. Now that I think about it, until the last few weeks I don't think it's ever locked on NOTHING before.

DCNUTTER, I can see where you're coming from with those tests, but in the real world the photos one wants to take often aren't of high contrast subjects or not in ideal lighting. I chose to upgrade from 400D to 7D because the focusing system is supposed to be so much better. I would really have to go back through lots of photos to check, but I don't think I had the same problem a few months ago.

james_so, I probably will take it in to have someone look at it, but after 9 months I'm sure I won't get a replacement.

Unfortunately I won't have time to do any controlled tests today, maybe some spur of the moment shots as was the case yesterday. But

kenskid asked if I could repeat the totally out of focus shot and I had surprisingly little difficulty doing that.

Tiny

Apple iPad mini 4

09/22/2011 09:03:11 AM · #73

Not that it is of any help but I just tried to shot with mine on AF and whatever I do the shutter will no operate until it has found focus.
I have tried all modes of focus with the same results.

Message edited by author 2011-09-22 09:03:22.