Is my Nikon 70-200mm f/2.8 sharp?

alexthecheese · 29 May 2012 at 17:49

Probably me just being stupid, but is this photo as sharp as you'd expect from a D5100 with a 70-200mm f/2.8?

Photo is 1/1000sec, f/2.8, ISO 100, VR Active, JPEG Fine, Highest resolution. About as sharp as it's going to get?

It kinds looks like there's a slight shadow to the left on the writing. :confused:

If you go to Flickr, zoom in and go to the original, then zoom in a bit on your browser, you can see what I'm on about.

DSC_1269 by Alex Bayes, on Flickr

Steve M · 29 May 2012 at 17:59

Looks OK to me for fully zoomed in and i'm guessing hand held?

An Exception · 29 May 2012 at 18:04

Are you using any UV filters etc. on the lens?

D.P. · 29 May 2012 at 18:05

Why was your VR on active? Active VR is for when you are on a moving car/boat/helicopter, were you? This may reduce sharpness or produce shadowing.

Furthermore, with a shutter-speed of 1/1000th second any kind of VR can potentially reduce sharpness. VR/IS/OS/ should always be switched off unless you need it to be on.

Thirdly, I assume this is a hand held image (if it was a on a tripod then VR should definitely be switched off). To really understand the sharpness of a lens we will need to see photos taken form a high quality tripod (not a cheap red snapper job but something with a proper ball head and legs that wont vibrate when the end of the lens is tapped), have mirror locked up (or exposure delay), a timed exposure or use a remote. Use live view to focus using maximum magnification to verify focus. Better yet, use live view then switch to manual focus and then take multiple exposure with slightly different focuses (ala focus stacking). Select the best photo form the set. That is the measure of the true lens sharpness when remove as many other factors as possible.

I would also shoot in RAW and check the sharpness in LR etc, also play with the USM sharpening. The AA filter will create a slight softness on high contrast edges, the USM can artificially restore the edge contrast.

Is this the VR 1 or VR 2 lens?

alexthecheese · 29 May 2012 at 18:06

Yeah there's a Hoya UV filter on it. Shall I try it with it off?

If it is just me being ultra picky I can live with that, I just wondered if it is as sharp as it should be.

An Exception · 29 May 2012 at 18:07

Yeh, looks like your getting some weird ghosting... (Hopefully due to the UV filter)

alexthecheese · 29 May 2012 at 18:14

I'll go and try and recreate that shot without the filter on.

alexthecheese · 29 May 2012 at 18:28

OK, I actually think it's better. No VR, handheld at 1/500sec, ISO 200, f/2.8:

DSC_1539 by Alex Bayes, on Flickr

Clear grate by Alex Bayes, on Flickr

You can actually see it's a clear grate over their flue now, rather than it just being a messy blur. Could actually be the VR that's mucking it up though, I'll have to test that as well.

Are Hoya UV filters any good? It came with the lens.

At least I wasn't going mad.

EDIT: You can see how much clearer the little wooden post at the top end of the shed is even from the zoomed out pictures as well.

An Exception · 29 May 2012 at 18:38

Hoya are good, especially the pro 1 filters. But if you use filters, there is always a risk of ghosting.

I doubt the culprit is the VR, test it again with the VR. Also do a test at an angle and focus on some writing. This will allow you to identify the area of maximum sharpness, and if that looks like the above, then the lens doesn't need any calibration etc.

Tbh, I think the lens is sharp enough, especially on a 16mp crop sensor.

DJ_Switch · 29 May 2012 at 18:42

If you are wanting to check the sharpness of the lens then you'd be better to check it where you have focused and not checking the sharpness of a background element when shooting at f/2.8 at 200mm. It is not likely to appear sharp regardless of other factors which could contribute to image degradation unless I'm off here and you are actually focussing on the shed?

With that lens I'd make sure you've got the hood on though.

alexthecheese · 29 May 2012 at 18:45

No, focussing right on the writing of the van, leaning out of the window from our upstairs bedroom.

I'll try the VR on and off also, with the UV filter on and off and then post it.

I've just been sad enough to do a comparison gif of before and after, and it's actually really quite noticeable (when you compare them directly

) how much sharper the 2nd is, without the UV filter:

The gif animator has knackered the quality of the photos but you can still tell, particularly with that flue thing.

Razor-BladE · 29 May 2012 at 18:51

You lose some contrast with the filter as well.

I might do a test at some point as well. Have a Kenko Pro1 filter on my 10-22mm when in a club, wonder how much sharpness I lose from that..

D.P. · 29 May 2012 at 19:45

An Exception said:
Yeh, looks like your getting some weird ghosting... (Hopefully due to the UV filter)

I think that is due to the VR, it will make a lot of ghosting in out of focus areas.

Edit:
Check out this great website which explains and shows why image stabilization will negatively affect out of focus areas of an image.

You also have to be very careful using IS/VR when you have fast shutter speeds or when you have a stable lens on a good tripod. My 70-200VR looks horrific if VR is left on when the lens is tripod mounted on a Gitzo 3531 + BH55

ShakenNstirred · 29 May 2012 at 19:51

VR shouldnt be the problem if your hend held
as it cuts out shake from you when at high zoom

if your on a tripod then yeah vr should be off

D.P. · 29 May 2012 at 19:56

ShakenNstirred said:
VR shouldnt be the problem if your hend held
as it cuts out shake from you when at high zoom

if your on a tripod then yeah vr should be off

If the shutter speed is very high then VR should be turned off. The gyros and the motor have difficulties perceiving and reacting to camera shake over such a short period of time and can induce blur.

Furthermore, as explained above, VR/IS can make harsh outlining and ghosting in out of focus areas so should always be switched off unless needed.

alexthecheese · 29 May 2012 at 20:53

It really doesn't help that I did the comparison by changing two factors does it!

Where's the great website about using VR, DP?

As a rule, when should I stop/start using VR? 1/1000 is obvious to me now that it's not necessary for VR. Am I looking at it down at 1/100 and lower? Ish? Suppose it also depends on the focal length.

D.P. · 29 May 2012 at 20:58

oops:

http://www.bokehtests.com/Site/Stabilization_and_Bokeh.html

When you use VR is dictated by the focal length of the lens, if you have access to a tripod, how much you care about sensor noise and what ISO you are at.

At 200mm and given that I didn't have a tripod for some reason I would use VR on shutter speeds of 1/250s or slower. If there was any motion in the scene then I would simply up the ISO.

alexthecheese · 29 May 2012 at 21:14

Cheers DP, I'll read that.

I can't test it properly 'cos it's getting lower in light, but it looks like it's the VR that's doing it (I've got the VR1 lens, btw DP).

I've just tried it without VR, handheld, and I can get as much detail on that flue, without the ghosting, with the UV filter on and the VR off.

I've also tried it with the VR on and there's no ghosting there, but the image is clear at only 1/30ths shutter speed.

I'll try again tomorrow but it does look like a case of the VR smudging my otherwise perfectly good shots. Sorry.