Nikon 24-70mm woes :(

Status
Not open for further replies.
sammyd56 said:
OK, for the interest of those following the thread, here are some better samples. Both shots at f2.8, 1/640, ISO 400, focus on the silver thing in the centre. EXIF should be intact. Both SOOC, unsharpened and taken minutes apart on the same D90 (so shooting the 24-70 in it's 'sweet spot'). Check out that bottom-left corner :eek:

http://www.samdobsonphotography.com/nikon2470.jpg

http://www.samdobsonphotography.com/tamron1750.jpg
Click to expand...

That's not really a valid comparison - Nikon shot at 70mm f/2.8 will have different DOF than Tamron shot at 50mm f/2.8. Your left corner softness therefore is simply because it goes out of focus - this can actually be seen if you scan the whole frame across - right side is sharper simply because it is closer. This can be emphasized a little if the lens is front focusing a bit.
 
You test images are worse than useless. Have you tried 'normal' pictures? If you must do these kind of tests - make the conditions more controlled - and shoot flat subjects straight on.
 
sammyd56 said:
One more sample then, I won't post any more comparisons because they won't be exactly the same framing as I didn't use a tripod... so tell me, does this really look normal? f4 this time, 1/128, on the D700. Same deal as before, SOOC unsharpened.

http://www.samdobsonphotography.com/2470-6.jpg
Click to expand...

It does actually look ok - perfectly sharp in the middle. The sides are difficult to judge as you shot at a strange angle so they are out of DOF areas. If you want to establish the sharpness/focusing issues then you need to shoot so that the areas of interest (newspaper, test chart, house wall etc) are all in DOF and using weird angles won't help to achieve it.
 
I'm wondering if Fixation will come back with the response that the lens is fine, it's the user that is at fault. The user seems to be confusing soft parts of the images with out of focus due to DoF. Despite several requests the user hasn't been able to come up with a controlled test to show the sharpness. The wall shots look fine to me in the centre, of course the plants at the bottom will be soft because they aren't the focal point.

It shocks me how people can buy about £2,500 worth of kit and not know how to shoot a simple photo to test the sharpness of it.
 
Judging by the "test shots" taken at strange angles of the subject and narrow dof at these distances and large apertures it is impossible to come to a conclusion as it look just like a dof issue to me.
 
Can't see much to complain about in those images myself sammyd56.

Why not stick enough newspaper over a wall to fill the frame, set the camera on a tripod square on to the newsprint and take pictures with your two lenses (at various apertures if you like) and compare?
 
Thanks to all those who have made suggestions, and extra thanks for those who did so politely. Will report back shortly :thumbs:

To the user in post #46, my name is Sam. Easy to throw around undue criticism when we de-personalise isn't it, but remember that internet communities do actually consist of real people. Cheers ;)
 
sammyd56 said:
To the user in post #46, my name is Sam. Easy to throw around undue criticism when we de-personalise isn't it, but remember that internet communities do actually consist of real people. Cheers ;)
Click to expand...
I don't think I did de-personalise it. I may not have used your name but my post made it perfectly clear that it was aimed at you, I was just too lazy to remind myself of your name as I typed my post. My name is Rob by the way, nice to meet you. :)
 
These latest shots are no good either as has been pointed out to you...
Please: shoot flat-on, tripod-mounted to eliminate any camera-shake, lock up the mirror if possible and take alternate shots with AF and manual focus at all the different apertures to eliminate that as a possible source of softness...

Everything you've shown us so far looks like focus errors or camera shake.
 
It was indeed quite clear that your entire post, including the last line, was aimed at me. That's why I hinted you could perhaps refer to be my name, which is in the post right above yours, just over there on the left... then you might think a bit more next time before posting comments that serve only to aggravate. Or maybe not :lol:

Nice to meet you though :)
 
sammyd56 said:
It was indeed quite clear that your entire post, including the last line, was aimed at me. That's why I hinted you could perhaps refer to be my name, which is in the post right above yours, just over there on the left...
Click to expand...
Actually, to be fair to my lazyness, I posted at post #46 however since this thread rolled over to page 2 at post #41 you hadn't posted so although your username was visible when quoted by other people, your actual avatar and name weren't to be seen on the page I was posting on.

Anyway, I assume you've dropped your lens off with Fixation now so lets hope for a quick turnaround and a cheap solution to your problem because the 24-70mm is a cracker of a lens and one you should love using (which you clearly don't at the moment). Good luck with getting it resolved. :thumbs:
 
Rob, point remains, the rude comment was not welcome. Am going to shoot a test shot as advised and if the consensus is that the performance is fine I will be happy. I'm not above admitting user error, we're all human. :)
 
There's a very interesting, useful and informative article on how to test lenses properly in AP, cover date 8th May (back issues are available). It advises (among other things) that the camera/subject distance needs to be 30x to 120x the focal length of the lens being tested and that everything needs to be dead square (tripod mounted etc). A couple of shots at a maginal shutter speed for hand held exposures is probably not an adequately accurate test.
 
sammyd56 said:
As you wish Mr. Moderator, I wouldn't want to deprive people of knowing the outcome. Will be back shortly with a proper test shot.
Click to expand...


I thought that the lens was bein takin down to fixation (as post #22) :shrug:
 
Nod said:
There's a very interesting, useful and informative article on how to test lenses properly in AP, cover date 8th May (back issues are available). It advises (among other things) that the camera/subject distance needs to be 30x to 120x the focal length of the lens being tested and that everything needs to be dead square (tripod mounted etc). A couple of shots at a maginal shutter speed for hand held exposures is probably not an adequately accurate test.
Click to expand...

You are right the tests should be done square on especially at close range and at F2.8. The two examples of the wall are a completely unfair test. Firstly the tam was shot at 50mm and the Nikon shot at 70mm both are close to the wall and taken at an angle, the depth of field runs parallel to the focal/back plane of the camera. DOF at 10 feet is about .69ft at 70mm and jumps to 1.36 at 50mm. So giving the Tam a unfair advantage.

Go and re shoot both at 50mm and directly at the wall and I suspect we will see a different result,,,, personally I am not convinced there is an issue here based on the samples provided. I think it is the OP lack of understanding of DOF and how it works. I may be wrong, I often am :D

Looking forward to the re-test with both lens, shutter speed f stop and mounted on a tripod parallel to the wall.
 
sammyd56 said:
Right, I don't have a tripod with me unfortunately, nor any other lenses for comparison, so this isn't a perfect test but hopefully good enough to make a reasonable judgement.
Click to expand...
What do you think?
 
That does look soft - but remember what I said about checking the AF by shooting the test shots on manual focus?
That could just be OOF...D700 and D3/D3x have the means to fine-tune AF discrepencies and store that info so that it's applied to that specific lens each time it's fitted.

It really needs some other reference in the shot to guage if it's front- or back-focussing.

Optically it may be brilliant, but you'll never know it if the AF is out of wack...
 
I agree with Arkady it does look a little soft but there are so many variables to take into account, in camera adjustment is good advice, but if you are really unsure take the camera and lens and get it tested at fixation. They are pretty well built my first 24-70mm went careering across a concrete drive from height, worked perfectly after LOL, well until I dropped it a second time and smashed the glass:eek:
 
gpa said:
I agree with Arkady it does look a little soft but there are so many variables to take into account, in camera adjustment is good advice, but if you are really unsure take the camera and lens and get it tested at fixation. They are pretty well built my first 24-70mm went careering across a concrete drive from height, worked perfectly after LOL, well until I dropped it a second time and smashed the glass:eek:
Click to expand...

Is that the one you sold to a bloke who sold it to Sam? :p

Like everyone else has said, I dont think the test shots are that good as test shots. Shoot a portrait focussing on eyes.
 
cambsno said:
Is that the one you sold to a bloke who sold it to Sam? :p

Like everyone else has said, I dont think the test shots are that good as test shots. Shoot a portrait focussing on eyes.
Click to expand...

:D no it ended up in the Bin !
 
Sam, try printing out this focus chart, shooting it using a tripod and you'll see if there are any front or back focussing issues then you can use the AF micro adjust on the D700. Once you've got that sorted then you can move on to tests like you've already done to see if the lens is sharp or not.
 
Was just testing AF. Put the lens on the floor (no tripod atm), focused on subject, used AF, click. Fine. Switched lens to MF, put on Live View, manual focus zoom in, click. Subject is OOF, focus as moved behind it. 1/500. Camera didn't move between shots. Repeated this three times to make sure. Switching the AF switch to M on the body as well seems to cure it.

Serious question for people who have this lens, when you hold it horizontally and shake it, lightly, from side to side, can you hear anything moving?!
 
sammyd56 said:
Serious question for people who have this lens, when you hold it horizontally and shake it, lightly, from side to side, can you hear anything moving?!
Click to expand...
I don't own a 24-70 but I have used one and no, you shouldn't hear anything moving. None of my lenses make any noise when I shake them, not that I do ever shake them though.
 
parkmoy said:
:agree:
Click to expand...

People were advising to do more controlled tests and that seemed sensible before taking it to Fixation.

Thanks Rob, me neither in normal circumstances, but I can't understand how I can put the lens in MF, focus perfectly in live view, click the shutter and have an OOF photo. Until I turn the body on MF that is. Makes me think there is something funny going on, maybe something loose inside, so I gave it a light shake and can hear a clunking sound. Quiet, but it's there.
 
sammyd56 said:
Was just testing AF. Put the lens on the floor (no tripod atm), focused on subject, used AF, click. Fine. Switched lens to MF, put on Live View, manual focus zoom in, click. Subject is OOF, focus as moved behind it. 1/500. Camera didn't move between shots. Repeated this three times to make sure. Switching the AF switch to M on the body as well seems to cure it.

Serious question for people who have this lens, when you hold it horizontally and shake it, lightly, from side to side, can you hear anything moving?!
Click to expand...

Try the same brickwall test again, but this time use Liveview contrast detect AF. Then set the manually lens to MFD, and switch the lens to AF and take the same shot. Compare both. Do the same test twice. Post 200x200 center crops from both.

Please use a tripod so the camera doesn't move between shots - this is important!
 
Andy I'd love to but like I said I don't have a tripod here. The ground or a flat surface is the best I have right now and that isn't very helpful.
 
sammyd56 said:
Andy I'd love to but like I said I don't have a tripod here. The ground or a flat surface is the best I have right now and that isn't very helpful.
Click to expand...

As long as you don't lean forward or backwards between shots, try without... I'd like to see contrast detect Liveview AF vs phase detect AF.

Do 3 sets of each, and put up the sharpest of each.
 
sammyd56 said:
As I needed a flat surface I did it inside. 200x200 crop from the exact centre with the method described above. The first is contrast AF. Looks pretty much identical to me.

http://www.samdobsonphotography.com/200x200.jpg
Click to expand...

That's good means your D700 has spot on AF then.

A well adjusted camera should produce the same results here - but often don't.

Not sure either way on the lens. My gut feeling is that you don't really have a problem as it seems your Tamron is pretty much the same sharpness (within margin for error), and there really is not a lot of difference between Tamron and Nikon f/2.8 lens (on sharpness).. so if you have a bad Nikon you have a bad Tamron too, so it all seems a bit unlikely.

IMHO if you send it in, it'll come back as "within spec".
 
Status
Not open for further replies.
Back
Top