70-200 VR soft wide open?

[Mr_T]

Hello there people of Talk Photography, I am in need of your technical know how to see if there is a problem here or not. I just purchased this 70-200 VR lens second hand and I can't help thinking it looks a little soft wide open at 200mm. I have shown others some of whom have the same lens and they seem to think it looks a little bit soft too.

I have two examples here, the first one is at 200mm at F2.8 and the second one is at 200mm at F8. The camera was placed on the floor and I used shutter delay and a remote release to make sure the camera was completely still. These were all taken on a D200 at ISO 100.

http://img.phyrefile.com/dr_q/2008/12/24/_DSC0001.jpg

http://img.phyrefile.com/dr_q/2008/12/24/_DSC0002_000.jpg

So I guess the question is whether to send this back or not. :|

 

[Steveo_Hants]

Looks good to me.

 

[mole2k]

At f8 it's nice and sharp, I would agree that it's pretty soft at F2.8, i've not experience with the lens so i've no idea what to expect but I know the 70-200 f2.8's are sharper than that wide open.

 

[ruvor]

i've not experience with the lens so i've no idea what to expect but I know the 70-200 f2.8's are sharper than that wide open.

Bit of a wild contradiction surely?

 

[Mr_T]

Bit of a wild contradiction surely?

Both manufacturers have variants of this lens so it may be that mole2k is referring to a Canon version of the lens. In an ideal world I could get hold of a new one and compare it to that but I can't so I need to borrow other people's eyes here instead .

 

[ruvor]

you'll usually find that all lenses look crappy when shooting test shots of text under available (indoor) light - tripod, mirror up or whatever never seems to make much difference. if you really want to test it either wait for good light and take some real shots, or whack a flash on the camera and try some at 200mm, 1/250 f/2.8 vr off (no filter on lens).

 

[allanm]

As Ruvor says, take it outside and take some proper shots with it.
At f2.8, you will find it a little softer than f5.6 or f8, but it is still pretty sharp. As you close the aperture down, it will become even sharper.
The photos you took look pretty good to me, I would go out and use the lens to get used to it. I will upload some photos to my Flickr this evening ( link below) that i took with my 70-200vr at f2.8 and you can see what my results are like.
Forgot to say, make sure VR is off when using the lens on a tripod or table, you only need it when hand holding.
Allan

 

[mole2k]

Bit of a wild contradiction surely?

Sorry as somebody else pointed out I mean to say it seemed soft compared to what my experiences with the canon variant of the lens.

 

[Mr_T]

VR was completely off with both of these examples. I also used the shutter delay as some of the exposures were pretty lengthy as well as a remote release to try and take out any human or camera error.

 

[cannockwolf]

they look fine to me and at 2.8 there's not going to be a lot in focus anyway at that distance, i have constantly shot mine @ 2.8 in low light to get shutter speed and never been unhappy with the results

 

[HoppyUK]

Apart from observing that your trial pics don't look too bad for what they are, that is no test from which to draw meaningful conclusions.

The subject is too close, the depth of field too narrow on what is actually a curved surface, and the shutter speed too long.

You need a distant subject in bright light to eliminate potential errors. But all lenses have their limitations, and if you go pixel peeping with unrealistic subjects it's usually not hard to find them.

Richard.

 

[Mr_T]

The subject is too close, the depth of field too narrow on what is actually a curved surface, and the shutter speed too long.

Thanks for the reply, I appreciate it. As for the shutter speed being for too long both of these photographs were taken at the same time. The first one is 1/3 of a second and the second for 3 seconds on a hard floor using a remote release. I also used exposure delay mode to try and cut down on any movement inside the camera from the mirror.

It's probably not at all a fair test but I attached the 50mm f/1.8 to my camera and shot the same thing with an even shallower depth of field with the results turning out to be sharper. I know it's not fair putting a prime up against a telephoto however I still thought that the extremely shallow depth of field would somewhat take that out of the equation.

I also did another 'test' using flash to try and bring out any detail I could that wouldn't appear so much in the poor light but this didn't help. Maybe I'm just approaching this completely the wrong way. To be frank I'm a bit paranoid about it because the lens is worth so much more than the body I'm using.

 

[cannockwolf]

heres one on a dull day shot @ 2.8, i think tack sharp

one on a bright day

and 2 in dark/concert lighting

 

[cannockwolf]

when you start using it in anger you will love it trust me

 

[HoppyUK]

Mr T I understand your paranoia. I suffer it too

There are two questions: firstly, is this lens good enough for me? I strongly suspect that it is about as good as you'll get without going to primes, but that is one possible issue.

The second question: is this a good copy/sample of the lens? It's a top-end Nikon so you'd be very unlucky, but you might like to try this test which another poster found helpful. He's looking at a wide-zoom, but the principle is the same. I've just cut-and- pasted it from the other thread:

"Go for it leave1 Sigma is a good lens and you can't complain at that price. But try this test first, just in case it's been traded in as a duff copy, or been dropped or whatever.

Stick it on your camera at 10mm and f/2.8, say ISO 400 to get a guaranteed shake-free high shutter speed. A bit of noise won't hurt. Point it over the road at a reasonably distant subject to minimise focus inaccuracies, a shop sign/number plate is good, and focus carefully. Set exposure, lock everything on manual. Shoot four pics in quick succession so the light doesn't change, with the target subject in all four corners of the frame.

Take your card and pop it into one of those processing booth things, select the target subject in each frame, whack it up to max magnification and print them all out. A 6x4in section is perfect.

What you are looking for is four corner images that are EQUAL in sharpness. In truth, they probably won't look that great but this is a severe test and it's equality you're after, not ultimate sharpness. You will probably find that the four images are not quite the same, they rarely are even on the best lenses so don't panic unduly, but if one or more corners is obviously significantly worse than the others, the lens has been put together off-centre (the most common manufacturing fault) or has been knocked off-centre.

I say again that this is a severe test, pushing the lens very hard at its weakest settings. So why not also take another shot to see what it really can do. Set a mid-range zoom focal length, f/8, low ISO. Print that one 15x10in and prepare for a big smile

Cheers,

Richard.

Edited to add, if a lens is really duff, you don't even have to print the images out - just blow them up to max mag on the LCD screen on the camera and you'll be able to see if one corner shot is way different.