Is this sharp enough?

EMA747

EMA747

Messages
2,070
Name
Andy
Edit My Images
Yes
I would guess that as they are still cr2 files you've not applied any processing. This could explain any 'softness' that might be there
 
I can't seem to open your links but from what you've said I know without looking that they will show nothing relevant.

It is impossible to make any assessment of lens sharpness without having an identical side by side comparison with a lens of known good quality - same subject/time/place, same focal length, same camera setting and absolutely identical image processing. Honestly, anything less is meaningless.

But if you just want to be sure you've got a good copy, that's easy to test. I've posted a simple routine plenty of times before. On one of your recent threads IIRC :)
 
I do remember you posting a link to your routine. I'll give that a go.

I have no reason to think my lens is anything less than pin sharp. It's just that as I have spent lots of money on it I just want to check that it's ok before I get into using it a lot.
 
EMA747 said:
Hoppy I couldn't find it in one of my threads but do you mean the routine you described in this thread?
http://www.talkphotography.co.uk/forums/showthread.php?t=163142&highlight=sharp+copy

Should I do it at 70mm or 200mm btw?
Click to expand...

Yeah, that's the regular version ;)

No real need to test at different focal lengths - if an element or two has been assembled out of kilter it's out of kilter.

I guess that if a lens has been physically damaged then you might want to test at short, middle and long focal lengths, and also at different focusing distances, but you would almost certainly feel that kind of damage in the zoom and focus rings. And the IS would quite likely be playing up too.

You can test all these things until you're blue in the face and get in a terrible pickle with dozens of images to confuse you, especially if you go looking for trouble. But I recently bought the f/4 version of your lens and just did the basic 'good copy' routine to make sure it was okay. And guess what, it was! Entirely as you would expect. And you've got a warranty for peace of mind :)
 
natjag said:
well the white wall looks fine, I assume the darker wall is further back. at 200mm 5.6 thats roughly what I would expect.
Click to expand...

Yea the darker one is about 3 or 4 feet further back. They were just a quick out of my bedroom window shot. IS was on for both too.

So you think it's seems ok?

Cheers :thumbs:
 
HoppyUK said:
Yeah, that's the regular version ;)

No real need to test at different focal lengths - if an element or two has been assembled out of kilter it's out of kilter.

I guess that if a lens has been physically damaged then you might want to test at short, middle and long focal lengths, and also at different focusing distances, but you would almost certainly feel that kind of damage in the zoom and focus rings. And the IS would quite likely be playing up too.

You can test all these things until you're blue in the face and get in a terrible pickle with dozens of images to confuse you, especially if you go looking for trouble. But I recently bought the f/4 version of your lens and just did the basic 'good copy' routine to make sure it was okay. And guess what, it was! Entirely as you would expect. And you've got a warranty for peace of mind :)
Click to expand...

I will do your test tomorrow just to make sure and put my mind at ease. I assume I should do it with IS on and mode 1 right?

The warranty is the main reason I bought new not 2nd hand. That and Kerso's nice prices. :D
 
EMA747 said:
I will do your test tomorrow just to make sure and put my mind at ease. I assume I should do it with IS on and mode 1 right?

The warranty is the main reason I bought new not 2nd hand. That and Kerso's nice prices. :D
Click to expand...

If you're doing it hand held, then IS on. Do it in bright light - you still need a high shutter speed.
 
What is a minimum acceptable shutter speed for this test to have meaningful results?
 
EMA747 said:
What is a minimum acceptable shutter speed for this test to have meaningful results?
Click to expand...

1/500sec or faster, even with IS on - just to be sure. Basically, decent daylight but you'll be at f/2.8 anyway. Make sure the light doesn't change between shots, eg clouds.
 
I have tried shooting a page of newspaper stuck to a wall and come up with some strange results.

The image seems to be pretty much sharp or soft across the whole picture. What is strange though is that using the top and left hand focus points makes a MASSIVE difference to the sharpness of the whole image. Those shot with the other focus points have come out much softer but the aperture/shutter speed/light/ISO are all the same. Why would this happen? I assume it would be more likely a fault on the camera than the lens?

Any help much appreciated.

Andy S
 
EMA747 said:
I have tried shooting a page of newspaper stuck to a wall and come up with some strange results.

The image seems to be pretty much sharp or soft across the whole picture. What is strange though is that using the top and left hand focus points makes a MASSIVE difference to the sharpness of the whole image. Those shot with the other focus points have come out much softer but the aperture/shutter speed/light/ISO are all the same. Why would this happen? I assume it would be more likely a fault on the camera than the lens?

Any help much appreciated.

Andy S
Click to expand...

1) If you did what you said you were going to do, and had a go at the checks refered to above, you will have read in the linked thread: "shoot a distant target to minimise focusing errors and eliminate field curvature issues - something at the end of the street like a car number plate or road sign." A sheet of newspaper is not a distant target and it is also flat, and potentially problematic.

2) If you go looking for problems you will probably find them.

3) Post some pics and full info.
 
I was going to do no 1 but as my 400D doesn't have focus points at the edge of the frame it is quite hard to be 100% sure it is focus correctly with something right in the edge of the frame.
How much of the frame should a number plate fill for the test to work? ie how distant should it be.
 
EMA747 said:
I was going to do no 1 but as my 400D doesn't have focus points at the edge of the frame it is quite hard to be 100% sure it is focus correctly with something right in the edge of the frame.
How much of the frame should a number plate fill for the test to work? ie how distant should it be.
Click to expand...

Use centre point AF, focus and recompose. Size of target doesn't matter, so long as the camera gets a good lock on it and there's some clear detail to see.

Over ten metres should be okay. You just need to minimise field curvature and potential focus errors, and get a bit of depth of field which will be shallow af f/2.8.

You're doing a simple 'good copy' test, not seeking to examine ultimate sharpness. If you want to do that, you need to do an identical side by side comparison with another lens of known good quality.
 
You must log in or register to reply here.
Back
Top