Nikkor 50mm f/1.8G on D610 - back focus issue

Messages
63
Edit My Images
Yes
Upgraded my old D5100 to a D610 last week and noticed that most of my shots at wider apertures seemed to be out of focus. I did some tests last night with the camera on the tripod, angled about 30 degrees at a piece of printed A4 paper flat on a table about a meter or so away and the camera and lens combination is consistently focussing behind the selected AF point (center in this case). I can adjust with AF fine tune, but even at the maximum setting it's slightly better but still considerably off. It's definitely not camera shake because some text about an inch further up the page was nice and sharp but the text directly under the focus point was blurry.

The only other FX lens I have is my Sigma 105mm macro so I test that as well and it's fine. I did also test both those lenses on the old D5100, there's no adjustment there but both of them were better than the 50mm on the D610 so it seems to be an issue between the 50mm and D610 only.

What are my options here, is it the lens that might need some attention or the camera, or both? The camera was bought used from MPB.com so there's some warranty there, the lens I've had for a couple of years so I guess I'd have to foot the bill to get that sorted, if it comes to that.
 
bought used from so there's some warranty there


It would not be reasonable to make a warranty claim on such
a focus issue, I think. A simple check in the UM will tell you if,
in the AF menu, the AF Fine Tune setting is available on your
camera and how to use it.

I now have 5 bodies — just added the D850 — and all bodies
have to go through the AF Fine Tune on all my 16 lenses…

Tedious but necessary. Have a good time! :cool:
 
Thanks, although I did say that I already did the AF fine tune and even that doesn't fully correct the problem.
 
that doesn't fully correct the problem.


Well, are you sure you performed the adjustment
in the right conditions? …this is critical!
 
What would you consider to be the "right conditions"?


  • fully opened lens
  • target at 30x to 50x the focal length
  • test at closest focusing distance
These should give you good results, good luck!
 
OK, the target distance would be about right, not sure what you mean about closest focusing distance? The guides I read online suggested to stop the lens down one full stop from wide open, so I had the 50mm set to f/2.8 I'd have thought 1.8 would be too shallow a DoF to be usable?

In any case, if I set the AFFT to -20 it did improve the focus of the lettering directly under the AF point but only a little, the letters there still were not the most in-focus on the sheet. Setting to +20 (the wrong direction) it got even worse. The other lens nailed focus on the desired point on the sheet pretty well every time.
 
You should be wide open at 1.8. After all these are going to be your smallest margins for error when using out in the field.

As above you should be about 30x 50x your focal length for the testing distance. So 50mm id aim for a range of about 1.8m - 2.1m. Make sure your at the same height as the chart focusing point too. Measure to your sensor marking on the camera (line with a circle marking)

Which chart did you use? Link us.

Did you ensure that your focus point wasn't touching anything else on the chart i.e. you focus point should be in the center.
Focus points can be a little off in the camera, so if it's close to one of the markings it might of focus on something else.

hope that makes sense.
 
not sure what you mean about closest focusing distance?
Set — manually (AF off!) Y— the lens at the closest
focusing distance
The guides I read online suggested to stop the lens down one full stop from wide open, so I had the 50mm set to f/2.8
Your decision… as who's tip you want to follow!
I'd have thought 1.8 would be too shallow a DoF to be usable?
What do you want DoF for when shooting a 2D target?
it did improve the focus of the lettering directly under the AF point but only a little
On my cameras, I could go from -20 to +20 if needed.
If -2 is not enough, try -10 or anything… you cannot
dammage anything!
 
So this is the guide I followed - there is a mistake in the text, the author gets the + and - values the wrong way round but apart from correcting that, this is pretty much what I did:

https://digital-photography-school.com/how-to-do-autofocus-fine-tuning-on-your-nikon-dslr/

I found this via Google, it was just one of a few pages describing essentially the same process so I figured it was probably fairly reliable. I'm still fairly new to photography and it's the first time I've had to do this so I don't know any better than the information I can find!

@ Kodiak QC, so you mean manually defocus the lens to the minimum distance before allowing AF prior to taking the test shot? As for 2D subject, using the method I linked to above, the target is 3D so DoF comes into play.

I'll export the test shots from last night later on and post them for example's sake.
 
So this is the guide I followed - there is a mistake in the text, the author gets the + and - values the wrong way round but apart from correcting that, this is pretty much what I did:

https://digital-photography-school.com/how-to-do-autofocus-fine-tuning-on-your-nikon-dslr/

I found this via Google, it was just one of a few pages describing essentially the same process so I figured it was probably fairly reliable. I'm still fairly new to photography and it's the first time I've had to do this so I don't know any better than the information I can find!

@ Kodiak QC, so you mean manually defocus the lens to the minimum distance before allowing AF prior to taking the test shot? As for 2D subject, using the method I linked to above, the target is 3D so DoF comes into play.

I'll export the test shots from last night later on and post them for example's sake.
If it was back focusing you would need to go into the minus - as you did.
 
So this is the guide I followed - there is a mistake in the text, the author gets the + and - values the wrong way round but apart from correcting that, this is pretty much what I did:

https://digital-photography-school.com/how-to-do-autofocus-fine-tuning-on-your-nikon-dslr/

I found this via Google, it was just one of a few pages describing essentially the same process so I figured it was probably fairly reliable. I'm still fairly new to photography and it's the first time I've had to do this so I don't know any better than the information I can find!

@ Kodiak QC, so you mean manually defocus the lens to the minimum distance before allowing AF prior to taking the test shot? As for 2D subject, using the method I linked to above, the target is 3D so DoF comes into play.

I'll export the test shots from last night later on and post them for example's sake.
Did you use a magazine like that article? If so that could be your problem.

use this chart instead: http://regex.info/i/FocusChart-v1.0-gray05.gif

Set chart at an angle.
Focus on the center line, make sure your all level and parallel.

Full guide here:
http://regex.info/blog/photo-tech/focus-chart
 
Last edited:
the target is 3D so DoF


Sorry, USE a 2D target making sure you are as perfectly as
possible in 0° angles in all axises!

Since I have many lenses and bodies, I use this…
where the great little ruler at 45° indicates how
much is in front or back focus.

Datacolor-SpyderLensCal.jpg
 
Last edited:
Yeah, back focusing (actual in-focus point was text about 1 inch further from the camera), adjusted down to -20, text under the focus point was still soft. Slightly better than before but not perfect. The Sigma lens nailed it right on text under the AF point every time.

No, not a magazine - an A4 sheet with some printed text on it, not as densely laid out as a magazine. The text under the AF point was the number 4 in a square box with a couple of lines' worth of white space above and below on the sheet.
 
It was done with artificial light, yes. Just read the first page of that thread (thanks for that, will read properly later), the person that did the test shots with the batteries was able to compensate using AFFT and get the correct battery in focus - I can't even get close at the maximum adjustment.

I'll try with natural light at the weekend, it's dark outside long before I get home from work right now...
 
I have had two 50 1.8G’s and have found both to be most unsatisfactory on my D600.

I have also had two 85 1.8G’s and found both to be unsatisfactory requiring more adjustment than -20. That was on a D750 and a D600.

Am I just unlucky or is Nikon QC poor?

The 50 1.8G also suffers from focus shift when you stop it down meaning that F4 can be far worse than F1.8. This focus shift has been referenced in a number of reviews.

My advice is as follows. Spend £40 on Focal software and let it do the hard work for you. If your still having issues and I was invest in Tamron/Sigma lenses that can be adjusted properly using their respective docks.

I took my issue slightly to the extreme and bought a 24-70 G2 and spent 8-10 hours (I know...) adjusting it properly. It’s a far sharper lens than the Nikon primes are because the primes were nearly always out of focus!!!
 
Last edited:
@markgodley, thanks for that, I'll have a read on that one. @Kodiak, that device looks useful, I guess it's not cheap though?

The LensCal is a good target, but you don't need to spend a bean. You can replicate that device exactly with a box of cornflakes and a ruler propped up against the side. Then set the camera up on a tripod square to the box and focus on it with centre AF point somewhere close to the ruler. Take a shot at lowest f/number, then check where the actual sharpest zone is on the ruler - just zoom in on the LCD to see. Adjust until it's right.

The keys to accurate AF microadjustment are:
a) a good target, flat and square-on, with clear, bold graphics for the AF system to lock on to, where there is no possibility that the camera can focus on anything else
b) a 'sensible' test distance - you can't put a number on it like a focal length multiplier (varies with format, and with application). In your case, try something between 2-3m

Even when you've got that right, you may still find focus is not dead accurate at all distances, and most fast lenses suffer slight focus shift at mid-range f/numbers. But don't worry about that, in practise it'll be pretty damn close and there are several other factors that are more likely to throw things out (down to you and technique).
 
The LensCal is a good target, but you don't need to spend a bean. You can replicate that device exactly with a box of cornflakes and a ruler propped up against the side. Then set the camera up on a tripod square to the box and focus on it with centre AF point somewhere close to the ruler. Take a shot at lowest f/number, then check where the actual sharpest zone is on the ruler - just zoom in on the LCD to see. Adjust until it's right.

The keys to accurate AF microadjustment are:
a) a good target, flat and square-on, with clear, bold graphics for the AF system to lock on to, where there is no possibility that the camera can focus on anything else
b) a 'sensible' test distance - you can't put a number on it like a focal length multiplier (varies with format, and with application). In your case, try something between 2-3m

Even when you've got that right, you may still find focus is not dead accurate at all distances, and most fast lenses suffer slight focus shift at mid-range f/numbers. But don't worry about that, in practise it'll be pretty damn close and there are several other factors that are more likely to throw things out (down to you and technique).

Just a bit more on this. You don't actually need a tripod, so long as the shutter speed is fast enough to avoid any camera-shake (use IS/VR). So...

a) A good target in decent light, flat and square-on, with clear, bold graphics for the AF system to lock on to, and where there is no possibility that the camera can focus on anything else.
b) Hand-holding is fine with a fast enough shutter speed. Use image stabilisation, and do not move between focusing and releasing the shutter.
c) An angled component, ie the ruler, so you can see easily where the point of sharpest focus actually is - in front, behind, or spot-on. It can be very hard and frustrating to assess the sharpest setting without that.
d) A 'sensible' test distance, similar to real world use. The temptation is to be too close as it's much easier to see what's going on, but then there's a danger that things will be fractionally out at longer range. You can't put a number on it like a focal length multiplier (varies with format, and with application). In your case with a 50/1.8, try something between 2-3m.
e) Longer lenses need greater distance, so choose a fence or similar, maybe a road sign, and shoot at an angle along it. Focus on a post and check the sharpest zone on the fence either side (or grass etc).

HTH :)
 
Last edited:
have you tried the lens on another camera body to see if it has the same problem ?
 
I did, as in the original post, I tried it on my older D5100. It didn't consistently back focus on that but I've always found the AF on the D5100 to be pretty variable anyway (one of the reasons I upgraded), the test shots I did showed the in-focus points to be a bit all over the place.

I'll try and get time for some better testing as per the methods in previous posts this weekend.
 
The good thing about Focal software is it tells you what af number adjustment is required.

For example it does the tests for you and then says set -7.

So much easier that guess work involved with the testing methods mentioned above. I have wasted so much time on the above mentioned tests with variable results and still no clear idea of what number to set the camera too.
 
Thanks for that, I'll look into that. Of the two versions available, is there enough benefit to warrant the more expensive version, for a keen amateur who probably won't have a large number of lenses?
 
Thanks for that, I'll look into that. Of the two versions available, is there enough benefit to warrant the more expensive version, for a keen amateur who probably won't have a large number of lenses?

Start with the basic version. You can upgrade later for no financial penalty should you wish to.
 
Back
Top