Am I expecting too much from autofocus with Nikkor 35 1.8

[gljivi]

I recently bought Nikkor 35 DX 1.8 and I'm using it with my Nikon D7000. It's generally fine lens, but for a last few days I can't decide is autofocus with this lens acceptable or not? I have a feeling that it should be better, but I never used this lens before. Maybe I'm spoiled with my old Nikkor 50 D 1.8, which is somehow always dead-on. I'm talking about focusing with f 1.8, to be clear.

I tried with fine tuning, but results are just not consistent, it will focus great for a few pics with lets say -5 for a target few feet away, then it will miss with more distant target, or it will backfous on close target and then front focus on distant target. And next day it will be fine with -10 on close target but for distant targets it will be best with fine tuning completely off. Problem is that it just happens however I adjust fine tuning, so it's just not consistent.

Here are two example pics. In first example box behind is little more in focus, and the first one should be more in focus. This pile of garbage was around 7 meters (20 feet) from my camera. On second try focus was ok, and then on third like this again.

In another example shoes are around 8-10 feet away and focus should be on second one but it's obviously more on first one. But then on second and third try focus was just fine. and then on fourth it was like this again.

So, I want to ask, is this kind of missed focus, as you see on this two pics, too much or should I accept that it happens with f 1.8 around 30 % of time? Am I nitpicking with this or this lens should autofocus more consistent?

I took this pics with AF-S and only central focus point selected, no auto anything.

Thanks for your patience and help

 

[ruifo]

Welcome to the forum.

I believe this is not a focus issue, but a "D7000+35mm @ f/1.8" combination shapness issue.

DxOMark website shows that at f/1.8, this lens with this camera (D7000) generates a quite soft image:
Tests and reviews for the lens Nikon AF-S DX Nikkor 35mm f/1.8G mounted on Nikon D7000 Measurements - DxOMark

On the D5300, for instance, the same lens at f/1.8 generates way better results:
Tests and reviews for the lens Nikon AF-S DX Nikkor 35mm f/1.8G mounted on Nikon D5300 Measurements - DxOMark

My learning is that it's not whether this or that lens is good or bad, but however, if that specific "body+lens" combination works good for your needs or not. Review the links above fr further details. Click at "Sharpness", and then at "Field Map". Next select the f/stop of your choice for review.

 

[gljivi]

Welcome to the forum.

I believe this is not a focus issue, but a "D7000+35mm @ f/1.8" combination shapness issue.

DxOMark website shows that at f/1.8, this lens with this camera (D7000) generates a quite soft image:
Tests and reviews for the lens Nikon AF-S DX Nikkor 35mm f/1.8G mounted on Nikon D7000 Measurements - DxOMark

On the D5300, for instance, the same lens at f/1.8 generates way better results:
Tests and reviews for the lens Nikon AF-S DX Nikkor 35mm f/1.8G mounted on Nikon D5300 Measurements - DxOMark

My learning is that it's not whether this or that lens is good or bad, but however, if that specific "body+lens" combination works good for your needs or not. Review the links above fr further details. Click at "Sharpness", and then at "Field Map". Next select the f/stop of your choice for review.

thanks for your welcome, I'm happy to join this forum
On the subject, I see what you mean, but how that can be an issue if some part of the image is always sharp enough (for my expectations), the thing is lot of times it's part of the image inch or two behind or in front of target if I shoot from 3-4 feet?

 

[KmH]

Understanding Depth of Field in Photography

Shooting close, like from 3 or 4 feet can make DoF very shallow.

 

[jaomul]

Welcome to tpf. I suggest you try your lens at about F4 for a while and see how it does. If it is ok try again at F1.8. Remember also that the lens needs an area of contrast to focus. Trying on something blank may not suffice. Also many people focus and recompose. At f1.8 this can cause issues due dof

 

[gljivi]

Understanding Depth of Field in Photography

Shooting close, like from 3 or 4 feet can make DoF very shallow.

I understand this. Thats why I'm asking, am I expecting too much? As I said, maybe I'm spoiled with always perfect focus with 50 1.8. Let me give you example. two shots from tripod with 35 1.8 and two shots with 50 1.8. I know it's not exactly scientific, but this kind of results are repeatable, during daylight, on close distance, on long distance, in the dark street... 35 1.8 almost always, no matter how much I'm trying to fine-tune it, focuses little bit behind or little bit in front of actual focus point, while 50 1.8 on the same camera is always perfect.

First two pics are 35 1.8 lens. On first one focus is on 106-107, and it should be on 110. So it's backfocusing. On second one it should be on 91, but it's on 95. So it's front focusing. Both pics are taken with exactly same settings two seconds apart. On two pics with 50 1.8 focus is mostly where it should be.

So lets say, for the sake of argument, that I have perfect shooting tehnique (I don't, but I'm not exactly complete noob), is this kind of variations acceptable?

 

[Derrel]

Those focus tests show NORMAL, absolutely EXPECTED minor differences in the depth of field. There is only one,single actual distance that is in perfect focus; every other distance is either in front of or behind, that specific distance. Depth of field is about what appears "acceptably sharp" in a print viewed at what is called appropriate viewing distance. When shooting photos like your test shots, it is ***expected** that one, exact, specific distance will be the sharpest; and everything else will be a bit less sharp at close to the focus distance, and more and more un-sharp as the distance away from focus distance grows. When we look at BIG images, we can see minor focusing issues; when a picture is reduced to 1200 pixels wide for the web, or printed as a 4x6 print, a lot of the issues "go away".

One issue with a 35mm lens is that the actual place the focus bracket indicator covers is relatively large, in the real world. At say 10 feet, the focus bracket covers an area larger than when a longer lens is used. With a 35mm lens, the AF bracket is not a "laser sight"...it actually covers a small area of the real world, so aiming it is not super-precise...it's a good guideline, but again, it is NOT a "laser-sight" type of aiming device, so focus errors can and will happen.

Questions about fundamental AF issues like your, a a HUGE part of TPF and other on-line forums.

 

[sk66]

AFMA is really only suited for one distance. I.e. if you fine tune for short distances it can make AF at long distances worse.

 

[gljivi]

Those focus tests show NORMAL, absolutely EXPECTED minor differences in the depth of field. There is only one,single actual distance that is in perfect focus; every other distance is either in front of or behind, that specific distance. Depth of field is about what appears "acceptably sharp" in a print viewed at what is called appropriate viewing distance. When shooting photos like your test shots, it is ***expected** that one, exact, specific distance will be the sharpest; and everything else will be a bit less sharp at close to the focus distance, and more and more un-sharp as the distance away from focus distance grows. When we look at BIG images, we can see minor focusing issues; when a picture is reduced to 1200 pixels wide for the web, or printed as a 4x6 print, a lot of the issues "go away".

One issue with a 35mm lens is that the actual place the focus bracket indicator covers is relatively large, in the real world. At say 10 feet, the focus bracket covers an area larger than when a longer lens is used. With a 35mm lens, the AF bracket is not a "laser sight"...it actually covers a small area of the real world, so aiming it is not super-precise...it's a good guideline, but again, it is NOT a "laser-sight" type of aiming device, so focus errors can and will happen.

Questions about fundamental AF issues like your, a a HUGE part of TPF and other on-line forums.

Thank you for the informative reply. I know it's not such a big variation. I'm asking, since I have no experience with this particular lens, is it expected and normal or I bought a lens thats maybe on upper limit of Nikon factory tolerances in precision of autofocus.

 

[gljivi]

AFMA is really only suited for one distance. I.e. if you fine tune for short distances it can make AF at long distances worse.

I noticed that, when I was happy with -5 finetuning for close objects, I noticed that anything further away was now in frontfocus. But then, next day it was the opposite, close objects were good with +5 and distant objects with -5 Thats why I'm wondering do I have a bad copy of this lens.

 

[Derrel]

AFMA is really only suited for one distance. I.e. if you fine tune for short distances it can make AF at long distances worse.

I noticed that, when I was happy with -5 finetuning for close objects, I noticed that anything further away was now in frontfocus. But then, next day it was the opposite, close objects were good with +5 and distant objects with -5 Thats why I'm wondering do I have a bad copy of this lens.

Well...there's also yet another possibility; the WAY you are actually USING the equipment has MORE error or more "variation" or "less precision" than the equipment itself! I am not trying to be rude, just explaining some things that I've learned over the years. For example, the AF brackets in the viewfinder--those often do not directly, 100% correspond perfectly with what the AF sensors "see".

The wider the field of view of the lens, the wider the area the AF focus bracket encompasses, which can lead to minor errors...or even significant errors.

The shorter the lens, the greater the depth of field; greater depth of field provides LESS-distinct cutoff between what the phase detection system calls "IN-focus" and what it considers to be "OUT-of focus". Determining precisely where the IN-focus and OUT-of-focus lies is the way phase detect AF works...and wide-angle lenses provide rather fuzzy end-points... so... YES< in a way, you are indeed expecting too much; this is not laser-like precision we are dealing with, nor laser-like accuracy. THere IS INDEED some actual, expected,allowed variation, and also the minor variations your test samples show seem quite acceptable to me. This is the real world...we're not dealing with micro-surgery here...

 

[gljivi]

AFMA is really only suited for one distance. I.e. if you fine tune for short distances it can make AF at long distances worse.

I noticed that, when I was happy with -5 finetuning for close objects, I noticed that anything further away was now in frontfocus. But then, next day it was the opposite, close objects were good with +5 and distant objects with -5 Thats why I'm wondering do I have a bad copy of this lens.

Well...there's also yet another possibility; the WAY you are actually USING the equipment has MORE error or more "variation" or "less precision" than the equipment itself! I am not trying to be rude, just explaining some things that I've learned over the years. For example, the AF brackets in the viewfinder--those often do not directly, 100% correspond perfectly with what the AF sensors "see".

The wider the field of view of the lens, the wider the area the AF focus bracket encompasses, which can lead to minor errors...or even significant errors.

The shorter the lens, the greater the depth of field; greater depth of field provides LESS-distinct cutoff between what the phase detection system calls "IN-focus" and what it considers to be "OUT-of focus". Determining precisely where the IN-focus and OUT-of-focus lies is the way phase detect AF works...and wide-angle lenses provide rather fuzzy end-points... so... YES< in a way, you are indeed expecting too much; this is not laser-like precision we are dealing with, nor laser-like accuracy. THere IS INDEED some actual, expected,allowed variation, and also the minor variations your test samples show seem quite acceptable to me. This is the real world...we're not dealing with micro-surgery here...

I'm not considering you rude, it's real possibility, that I must learn to use this lens properly. Only other lens with f1.8 that I have is this Nikkor 50 1.8 D, my other lenses are Nikkor 18-200, Sigma 10-20 and some telephotos, so this 50 1.8 is my only reference (and you must admit that those two examples taken with it are - almost - laser precision

 

[ruifo]

I don't know if that helps... But here are some shots I made with with ths D5200 body and the Nikkor 35mm f/1.8 lens, at 1.8. The focus was right where I wanted it to be. Click to enlarge.









Source:
https://www.flickr.com/photos/ruimc/sets/72157641188862373/

 

[gljivi]

Nice pics Ruifo! That first pic especially, I'm definitelly not sure could I autofocus on that flower and expect predictable results. I guess I'll just shoot some real life photos, not tape measure, for some time and then I'll decide what to do.