Am I being over critical?

Morph3ous

Morph3ous

Messages
598
Edit My Images
Yes
Since I got hold of my Canon 17-55, i've been fussing over whether its a sharp copy or not. I know I shouldn't, but I am, so I'm looking for some help.

I always thought i'd see a massive jump in quality when I upgraded from the Canon 400d/Sigma 18-200 to the Canon 60d/Canon 17-55, and I've been a little disappointed with some of the images taken so far. Its mostly ones of the children, shot wide open at f2.8 with the shutter at around 1/80. Something I thought the IS on this lens coupled with the 60d would handle easily.

As I'm not keen on posting photo's of the children onto public forums, I did a little test this morning as below.

Is this the quality you'd expect from this lens?

Shot handheld @ 1/160, f2.8, ISO200. AF hit on the focus limit switch centre of frame. Shot in RAW, exported to JPEG via LR3 with no sharpening.


Sharpness test by Minnis Photography, on Flickr


Sharpness test 100% crop.jpg by Minnis Photography, on Flickr
 
Looks slightly back focusing - the AF/MF is more in focus than the focus limit switch. Given it's hand held, it MAY be you have moved a centimetre towards the subject as you took the picture which would cause the appearance of back focus
 
The lens looks ok to me but without knowing your focus point the previous poster may be right.

Try taking longer length shots say outside on a tripod. A brick wall is good. Try at different aperures and focal lengths. Don't forget to turn off IS if using a tripod.
 
Yellowbelly said:
The lens looks ok to me but without knowing your focus point the previous poster may be right.
Click to expand...

It says in my first post that the AF hit on the centre of the image on the focus limiter switch.

I've just done the battery test to check the focus accuracy.
 
Morph3ous said:
It says in my first post that the AF hit on the centre of the image on the focus limiter switch.
Click to expand...
It's definitly focused behind that. Compare the paint texture around the switch with that on the main body. But as arad says that could be movement after focusing

Morph3ous said:
I've just done the battery test to check the focus accuracy.
Click to expand...
What's the battery test?
 
Battery test shows the focus is fine, hit the centre of 9 batteries and is in good focus.

Any other comments on the sharpness of the image would be appreciated.
 
gridlock said:
What's the battery test?
Click to expand...

Sorry, I didn't explain. Its the test where you line up batteries to check whether its front or back focussing.

A la...


Battery test by Minnis Photography, on Flickr

Seems okay to me. I set spot AF on the centre AF point, aiming at the 4th AA battery along (6th in total)
 
Last edited:
I was testing one of my lenses recently and thought it was front focussing. Did more tests and it seems ok, RAW files were softer than jpegs straight out of the camera.
 
Morph3ous said:
It says in my first post that the AF hit on the centre of the image on the focus limiter switch.

I've just done the battery test to check the focus accuracy.
Click to expand...
Sorry, I missed that bit.

All the "rough and ready" testing in the world won't tell you the final answer. Get out there and take some real world images, process them carefully then post us a few gems.
 
Morph3ous said:
Its mostly ones of the children, shot wide open at f2.8 with the shutter at around 1/80.
Click to expand...

Unless you have abnormally subdued children, 1/80 is going to be too slow to freeze movement.
 
squishy said:
Unless you have abnormally subdued children, 1/80 is going to be too slow to freeze movement.
Click to expand...

exactly my first thoughts too, no amount of IS can stop movement of the subject, it only takes fractions of movement to soften things wide open. I don't see any sharpness issues with the pictures you have psosted, so you need to either post a picture that displays the issue you are concerned about [or perhaps a less identifiable crop from it if you are so worried about your kids pictures] then we can possibly confirm better whether it is lens or user error. :)
 
Thanks for the replies.
 
Looks pretty sharp to me.
A technique for testing focus point I use is to open in photoshop desaturate the image and then emboss it I find it makes it much easier to see where the sharpest point is.

v9ri0.jpg
 
The problem with the battery pic is you don't know where it has locked focus. IF the lens is out, it is only out a teensy weensy (technical term there) bit. With a curved surface at a reasonable distance, you won't know if it is out. The old cereal packet square on plus a slanted ruler is the order of the day.
 
Just tried this test and was happy with the results...

IMG_0549.jpg
 
I think you're being way over-critical, I don't see much wrong with that shot of the lens at all. The very slight difference between sharpness at the intended point of focus and and just behind it is only visible in a 1:1 crop and could be down to a number of factors including user error. When you're making these sort of tests it's best to take several shots to establish a mean average although this sort of test shot isn't the best of subjects.

Your battery shot is a much better test and seems to indicate that all is well.

With regard to shots of the kids - you need a much faster shutter speed than 1/80th. Don't forget that IS can kill camera shake at that speed but it never ever does anything at all for subject movement. ;)
 
Your tests are not sufficiently accurate, and neither are any of the alternatives suggested if you want dependable results.

The most likely cause of the back-focus evident in the pic of the lens is the hand-held camera moving a few mms after focusing. Camera has got to be on a tripod, the target must be flat, and the distance 'realistic'.

With anything other than a flat target, ie curved or angled, you cannot be sure where the lens has actually focused, especially bearing in mind the AF point is quite a bit larger than the little square in the viewfinder. And testing at such a close distance is liable to throw up errors that are not visible at normal shooting range.

AF calibration is a compromise and absolute accuracy varies acording to both focal length and distance. If you push it right to the edge of the normal optimum operating envelope by shooting abnormally close, you run the risk of making a correction that will actually make the lens worse at normal distance.

That's just the question of AF accuracy. Testing for a 'good copy' is anothert question, and requires a whole different procedure.

When questions like this come up, nine times out of tens it turns out to be user error. Before you judge the lens, shoot some pictures outdoors, in good light, with careful focusing and high shutter speeds.
 
terryt said:
I was testing one of my lenses recently and thought it was front focussing. Did more tests and it seems ok, RAW files were softer than jpegs straight out of the camera.
Click to expand...

Raws will always be soft compared to jpegs straight out of the camera!
 
You must log in or register to reply here.
Back
Top