b top gun

Finally a break in the gloomy weather so I could set up a test of my copy of a Nikon 24-70 f/2.8 non VR lens. Primary reason for testing, check lens performance primarily at the 24mm setting from f/2.8 through f/16 in whole stops. Have questioned the image sharpness of the corners of this lens since I got it. Test target, my back fence, 5 foot high wooden privacy type with lots of detail; target chosen because of the detail available. Lens on my D850 on a tripod. Distance to the target, about 8 feet; D850 as square to the fence as possible; 2 second exposure delay selected to minimize camera shake. Shots were taken in sets of 6 exposures; one set in color, a second set in monochrome for extra detail. 24mm starting at f/2.8, about 65% of the image had acceptable sharpness from the center out. Did not get acceptable image sharpness until f/11 at 24mm and even then not sharp all the way to the corners/edges. Same lens, same string of shots except at 28mm; immediately better results. Did not shoot beyond 28mm by choice. All this with the camera and lens set for auto focus.

For comparison sake, switched to live view, full manual focusing, and used in camera focus peaking option available in the D850. Manual focus shots gave better results especially at the wider apertures almost immediately. From experience I know not to shoot this lens at 24mm and forget f/2.8 at pretty much all focal lengths. I start at 28mm, have learned the sweet spot of this particular lens is f/5.6 or smaller. When shooting landscapes and seascapes I have come to prefer f/6.3-f/7.1 minimum aperture and a tripod whenever convenient. Will try some 'scapes using live view next. I have been researching 24mm primes and lenses like the Nikon 14-24 f/2.8 to renew my faith in the 24mm focal length; want to use it for low light and night time shooting as well. All this for me begs the question...."Why would you make a lens with a max aperture of f/2.8 if said lens does not yield sharp images to at least 80% of the image from the center out???" Using my copy of the 24-70 f/2.8 is like shooting through a tunnel at times. One option is to select the 1.2 crop factor available in the D850 menu, however that crop factor computes to around 28mm soooooooo, just set the lens there and shoot away, which I have been doing. Another option is to put this lens on my D7100 and get something else wide angle for my D850. Note: I am not looking at third party lenses, Tamron in particular; have owned four Tamron lenses, returned two to B&H a week after getting them, dumped my 150-600 due to auto focus issues; still have the 18-270 but should have sent it back because it is softer than my Nikon 18-200 VR.

tradio Loc: Oxford, Ohio

I had to calibrate my 24-70 a few clicks to get it dialed in. It was front focusing I think. Have had no problems since.
If you do happen to look at third party lenses, I would check out the ART series from Sigma. I have a couple and they are outstanding in build quality and performance.

martinfisherphoto Loc: Lake Placid Florida

I've never met a wooden privacy fence that is perfectly plumb/flat. If you want to test lens at f/2.8 find something that is absolutely flat, maybe brick wall.

Julian Loc: Sarasota, FL

So much for pictures of your fence; but, have you tried to photograph everyday subjects like landscapes, people, animals, flowers? How do they look?

Kmgw9v Loc: Miami, Florida

In other words, just take some pictures.

Paul Diamond Loc: Atlanta, GA, USA

Hi, Agree with you. With a high MP camera like the D850, you need to look critically at the performance of your lenses.

I have not done straight forward tests of each FL/each f/ stop. And I should. Suggest you compare your earlier 24-70 to the VR versions of same FL lens to see if the newer designs perform better. Something like dxomark.com might help. I have a 24-70 VR Nikkor lens and found that the Tokina 24-70 was sharper, especially at the edges. And, I bought the Tamron 15-30 before my trip last year to Yosemite. I have been pleased with it's performance for a lens that I use rarely.

I need to set aside some time with a brick wall, tripod and an assortment of my lenses for similar tests.

sscnxy

You really should check your lens' AF for front or back focus and tune it if either is present. That won't address vignetting or loss of sharpness in the corners, but it will maximize sharpness at your presumed focus point.

larryepage Loc: North Texas area

If you look at the geometry of your test setup, you will note that when you are set up 8 feet from the fence, it is probably more like 12 feet to the parts of the fence at the corners of your image field. This lens is not corrected to focus on a flat field (like a macro lens most likely would be), and depth of field is not going to cover from 8-12 feet at f/2.8.

I'm not saying that there might not be a problem with your lens, or that it might not need focus calibration, but nothing that you do, nor any other lens not corrected to a flat field is going to get the entire field of that fence critically sharp at f/2.8. Wide angle photography presents many little challenges like this. Quite a bit of consideration is necessary to really do it correctly.

The improved performance at smaller apertures is completely expected in this case. Overall focus improves with increasing depth of field as you stop down. I have that same lens and suspect that its corners are fine.

Thomas902 Loc: Washington DC

top gun I would suggest you study larryepage post/wisdom carefully... Larry is spot on here.

The performance you are seeking is indeed available in micro nikkors which are primes and optimize for flat field rendering. The lens you are testing is designed for event shooters (think wedding here) where the big money is... Event shooters aren't really that concerned with corner acuity. Pick the right tool for the job, k?

However if you are going to stay the course and want to make the best of your 24-70mm...
"...Something like dxomark.com might help..." I agree with Paul Diamond here

Food for thought... If you are going to fine tune focus consider doing so at the Focal Length you will be using the optic at... anything else may be a waste of time/energy... experience is a brutal teacher.

EZ trick... place a mirror flush on the vertical surface you are using as a target... start LiveView and move your tripod until you see your lens in this center of the mirror... then place your target over the mirror.

Best Practices: Obtain either a D6 or the D780 which actually allows fine-tuning zoom optics at min/max... Canon users have enjoyed this feature for a dozen years... Nikon is a tad late to the dance...

Below is from a D6 review by a commercial shooter I have considerable confidence in (Steve Perry) who I recently discovered is also elated about Nikon's new Dual Fine Tune for Zoom Optics... I will be renting a D780 to see if it addresses a multitude of issues I have with my vintage "Pro" zoom optics...

Hope this helps...
All the best on your journey top gun...
.

(Download)

User ID

You asked why would they make such an (expensive) noticeably imperfect lens. The simplest answer is that at a marketing department meeting only one person spoke up for all the potential buyers who would shoot fences, square on, from 8 ft, at 24 mm at f:2.8, and that person was promptly reassigned to managing the cafeteria. Thus YOU had no representation at the meeting and so your needs were never considered.

Do you realize that at 8 ft, using a 24mm, your subject distance is only 2/3 of the 12 ft width of your subject field ? No expert tech sets up a copy camera at such a ridiculous ratio. Your results were doomed before you unfolded your tripod.

Another person at the meeting brought up something that you yourself discovered, that they’re marketing (expensive) 50MP cameras with a noticeably imperfect SLR AF system. That person recommended wasting no extra money perfecting an f:2.8 zoom destined for use on an AF SLR. That person earned a bonus check, was given a personal parking space, and was reassigned to the Z-series becuz he felt guilty about marketing AF SLRs as top grade devices. He reports that he sleeps much better now.

Thomas902 Loc: Washington DC

User ID makes good argument for upgrading to the NIKKOR Z 24-70mm f/2.8 S and a Z-7

Nikon would certainly like that...
So AF Fine Tune is then irrelevant thus no worries right?

Strangely, Nikon has included AF Fine-Tune in their Z6 and Z7 cameras – why is that?

b top gun

UHH never ceases to amaze me with all the free expert advice available; all ya have to do is post something and sit back and wait, they will find you. They also like spending your money for you.

User ID

Live view on-sensor focusing eliminates the usual manufacturing errors in the optical path to an SLR AF module. That issue is banished. But most current on-sensor AF systems incorporate PDAF as well as CDAF.

CDAF is pretty much foolproof but PDAF is faster. I suspect that certain lenses, possibly ultra wides or vast ratio zooms, may have peculiar imaging characteristics that cause PDAF errors. Something like that might be a reason to provide for AF fine adjustment. A bit of a mystery there.

I checked David Busch’s Z6 book and he has no clue, just vague conjecture, as to why a lens might not AF properly. Still remains a mystery.

Some Sony menus let you choose either PDAF or CDAF under “AF Mode”. I spoze they could relabel them as something like “Action” and “Still Life”.
_____________________________

BTW the Sonys that I use do have an AF adjustment menu but it’s grayed out unless you use an LAEA4 converter, which houses an SLR-type AF module for use with Maxxum/A-mount lenses. The LAEA4 instills SLR AF error into cameras that are otherwise immune to such problems. But it does allow AF with adapted SLR lenses, some of which are over 30 yrs old.

User ID

In this case the OP already uses a pricey top of the line lens and body combination which, despite correct use, unfortunately does a rather sorry job of photographing fences. Might just as well toss even more money at the problem.

When I shoot a wall, as I periodically do, I use a top of the line camera and a suitable lens, never a fast zoom.

SteveR Loc: Michigan

Best thing to shoot to test your lens is the brick on your house and, possibly, newsprint.