Is this lens as sharp as it should be? Possible front focus? The answer!

paulcamcas

paulcamcas

Messages
2,509
Name
Paul
Edit My Images
No
Anyone familiar with a Nikon 105mm f2 DC?

I've just bought a used one of these and I'm not familiar enough with the camera (D300s) or lens to be sure if it is a good copy and as sharp as it should be.

I've done a series of test shots on a tripod in good light, taken the same 600x209px centre section of each and them together into the image below. The images were raw + zero pp.

What do you think good or not?

Btw, it might be front focusing a tad.

battery-test.jpg
 
Last edited:
You've managed to focus between the batteries rather than actually on any of them, so impossible to say for the widest apertures, but stoppped down it looks as normal as you'd expect.
 
Fair point Alan. I think that means it is front focusing a tad because the words "Rechargeable NIMH" in green on white were central in the AF rectangle.
 
You would be better to focus on a series of flat items rather than round. Say a number of CF cards placed as you have with the batteries.
 
Also make sure you are far enough away to be well away from the min focus distance (whatever that is for this lens)
 
That is a heck of a good tip. Thanks Dale!

I've done bit more testing, this time with CF cards and it does seem to be front focussing by about the thickeness of a cf card. I'll try some fine tuning now and see if that does the trick.

Many thanks all.

Are the bits that are in focus as sharp as you would expect? This is supposed to be one of Nikons sharp lenses.
 
Okay, here's the AF test and set up. Focus was on the word "Hooray" on the flat surface. The number 16 on the ruler was in line. Looks like front focus to me.

test15.jpg


set_up.jpg
 
This is a bit of a blow. With maximum correction dialled in, the front focus is still over 1cm.

test29.jpg


test-2-emboss.jpg
 
Focus doesn't look like it's changed at all :thinking:

Your target is very close to the rule. The active AF area is usually larger than shown so it could be picking up on something else.

What distance are you at? Don't go too close and keep it realistic, or any correction you apply could throw it out at more normal shooting distances.

Edit: try a slightly different test. Don't use the rule and just shoot the flat target, dead square, and at a reasonable distance. Shoot right through from -20 to +20 at increments of 5. Zoom in on the LCD and flick through the images, which should all be slightly but noticeably different. Fine tune from there.
 
Last edited:
Thank you Hoppy and Scott. Sensor was 45 inches from the target. It was centre point focus.

I'll try focusing in the centre of the box so there is no chance it's focusing somewhere else.
 
Last edited:
Repeated test in line with advice: Further away (63 inches). focused on the word "is" in the centre of the box. Centre focus point.

test_3.jpg


test-3-emboss.jpg
 
Last edited:
HoppyUK said:
Edit: try a slightly different test. Don't use the rule and just shoot the flat target, dead square, and at a reasonable distance. Shoot right through from -20 to +20 at increments of 5. Zoom in on the LCD and flick through the images, which should all be slightly but noticeably different. Fine tune from there.
Click to expand...

Just seen this. More good advice Hoppy. And interesting results... did it again at 63 inches, centre point only, middle of the box, focused on the word "is" and found that there were clear differences with each 5-step adjustment and....

the sharpest rendition was with zero correction. :bang:

I reckon this means that all the other testing I did was flawed/subject to variation/rubbish.

I'm going to do some eyeball testing when my favourite model comes home. What counts is whether her front eye is in focus.

Many thanks for your help all and any further advice or suggestions are welcome.
 
Last edited:
paulcamcas said:
Just seen this. More good advice Hoppy. And interesting results... did it again at 63 inches, centre point only, middle of the box, focused on the word "is" and found that there were clear differences with each 5-step adjustment and....

the sharpest rendition was with zero correction. :bang:

I reckon this means that all the other testing I did was flawed/subject to variation/rubbish.

I'm going to do some eyeball testing when my favourite model comes home. What counts is whether her front eye is in focus.

Many thanks for your help all and any further advice or suggestions are welcome.
Click to expand...

Haha! Often happens ;) Good result though :)

You are familar with the problems of focus-recompose technique when shooting shallow DoF at close range? If not, or for anyone else that's interested, focus-recompose inevitably means moving/tilting the camera to obtain focus, then moving back to recompose the shot. In doing so, the focusing distance changes fractionally.

Solution is to use one of the outer focus points close to the eye, and don't move at all. You may have to move back a bit so you can crop to the desired composition later. (This will increase DoF, so keep that to a minimum.)

The other thing is that when DoF is measured in a few mms, even breathing, by the subject or photographer, will change things. Keep focusing and refocusing all the time. Some people use servo tracking for this.

Edit: be interesting to see how you get on with the DC feature. Sounds just the ticket :thumbs:

paulcamcas said:
Btw, please delete the apostrophe in "count's" in the quote, lord know how that got there :lol:
Click to expand...

LOL Done!
 
Last edited:
Yup, very familiar with focus-recompose problems. Part of my reason for switching to the D300s was to get focus points out nearer the outer thirds.

Btw, please delete the apostrophe in "count's" in the quote, lord know how that got there :lol:
 
HoppyUK said:
Edit: be interesting to see how you get on with the DC feature. Sounds just the ticket :thumbs:
Click to expand...

I'm very interested too. In choosing between the 85mm f1.4 and this lens, I was swayed by the price, extra length and chance to play with DC. Probably rubbish though, most gimmicks are. What I really wanted was a longish portrait lens with largish aperture and smallish price. This fit the bill perfectly.
 
paulcamcas said:
I'm very interested too. In choosing between the 85mm f1.4 and this lens, I was swayed by the price, extra length and chance to play with DC. Probably rubbish though, most gimmicks are. What I really wanted was a longish portrait lens with largish aperture and smallish price. This fit the bill perfectly.
Click to expand...

The dc isn't a gimmick - used well it gives a very subtle but very pleasing effect. Used on neutral it's a great portrait lens. Used badly you may as well by a soft focus filter
 
boyfalldown said:
The dc isn't a gimmick - used well it gives a very subtle but very pleasing effect. Used on neutral it's a great portrait lens. Used badly you may as well by a soft focus filter
Click to expand...

That's good to know Hugh. I'm still a bit worried about this copy. In a real world shot, it definitely seems to focus in front.

Here's a snap of lunch. The single AF point was slap in the centre of the salami but it's not the part that's in focus.

breadsalami.jpg
 
Last edited:
It does sound like that's the case - not that it should make a difference are you testing with the dc in neutral.

The only other thing I'd add although this is an ace lens I do get a lower keeper rate with this then anything else in my bag
 
Good point but, yes, DC locked in neutral.

I think I'm in trouble. I just took 20 shots at f2.8 of my wife at head-shot distance with single-point focus on the leading eye with the 105mm and in not a single one was that eye sharp. I did the same with my 17-55mm f2.8 and in every single one of the twenty that eye was sharp.
 
Thanks Hugh. I'm in Wolverhampton. I'll have to talk nicely to Fixation - I'm off to Italy on Thursday and I bought the lens especially for the trip so it'll be a real blow if I can't take it.
 
Can't you put that error right with AF microadjust?

Given your previous test results, I'm wondering if the focus is inconsistent. Did you de-focus everytime between shots? Have you done a sequence all at the same settings, de-focusing each time, to check consistency?

But I have to say a lens like that, of that quality, should be absolutely bang on. Hugh's experience is interesting. And that bread looks just delicious :)
 
HoppyUK said:
Can't you put that error right with AF microadjust?

Given your previous test results, I'm wondering if the focus is inconsistent. Did you de-focus everytime between shots? Have you done a sequence all at the same settings, de-focusing each time, to check consistency?

But I have to say a lens like that, of that quality, should be absolutely bang on. Hugh's experience is interesting. And that bread looks just delicious :)
Click to expand...

I'm going to try micro-adjusting again tomorrow. Not sure it will correct that much. I re-focused each time but didn't specifically defocus between shots. I'll add that to the mix tomorrow.

Agree that a lens like this should be bang on. It is in mint condition showing no signs of use or damage at all.

Lunch was very tasty, thank you :)
 
Last edited:
Yes, I tried manual focusing in Live view but I was worse than the AF. I'm going to try again tomorrow.
 
Update: Tried 10 shots with defocusing between each and every one missed focus.
 
paulcamcas said:
Update: Tried 10 shots with defocusing between each and every one missed focus.
Click to expand...

:(

Just had a google around and it seems these DC lenses have a bit of previous on front focusing. I can't believe it's anything that cannot be fixed though.
 
Last edited:
Yup, it's a blow. I'd rather return it than have the aggro and expense of getting it fixed but the seller has come over all incommunicado. Can't recall every having a problem with Canon kit and from what I've seen so far, the 135mm f/2 L knocks this lens into a cocked hat. My sojourn into Nikon-land is going to be a short one at this rate. (Yes, I am that fickle.)
 
Last edited:
Just to put the record straight, the chap I bought it from was away and is being stand-up about it. I'm going to have a chat with Nikon/Fixation tomorrow and either get it fixed or send it back.
 
Not sure if this helps, but, I don't believe lenses of the AF/AF-D variety can be corrected for front or back focus. Only AF-S lenses can be. Otherwise, it's all down to the body to focus. I reckon if you need extra 'af fine tune' ability, then its a fault with the body. I say this, as I had a D300 body that needed +20 to be even close to focussing. My D300s body used a mere +5 to sort it out. Same across the board with my other lenses that were AF/AF-D. It's the body that needs to be sorted.

One way to 'prove' this possibly is to use live view, and let it focus in live view. This uses contrast detection rather than phase detection auto focus, and I'm pretty sure that contrast detection, by its nature, will hit focus every time (or at least 100% of the time I've tried it) although it is a lot slower. But this will show you that the camera and the lens has the ability to focus properly themselves. Phase detect (I believe) relies on the body to communicate this info to the lens and so the body will need adjusting, rather than the lens, so I doubt the fault lies with the seller and the lens...
 
Very helpful Cheng, thank you. I don't understand everything you said yet, but I'll work on it :)

I did a head-to-head normal phase AF vs live-view contrast AF and it shows that the body/lens combo can get a plane into focus.

The body focuses accurately every time with the 17-55mm lens using normal phase AF. So why does this mean that it is the body that needs adjusting rather than the lens?

Here are the test photos: normal phase AF on left, live-view contrast af on right.

test-4.jpg


test-4-emboss.jpg
 
After much trialling of different bodies and cameras, the conclusion is that Cheng was absolutely on the money.

1. The lens worked perfectly on every other body we tried it on.

2. Every other lens works perfectly on my body, except for other examples of the 105mm f2 DC which all front focused like crazy.

It seem that my body just does not want to play with this particular lens model.

While I was at it, I tried a Micro-Nikkor AF-S 105mm f/2.8G IF-ED VR. Blimey, what a lens for 600 quid! So sharp. As soon as I get back from my hols I think I'll be investing in one of those.

Thank you all for your input.
 
paulcamcas said:
Update: Tried 10 shots with defocusing between each and every one missed focus.
Click to expand...

paulcamcas said:
2. Every other lens works perfectly on my body, except for other examples of the 105mm f2 DC which all front focused like crazy.

It seem that my body just does not want to play with this particular lens model.
Click to expand...

I'm not sure if I'm reading you right, but does this mean you're focusing, then adjusting the defocus control. If so that will explain the front focus issues you are having as you need to focus after you've adjusted not before
 
boyfalldown said:
I'm not sure if I'm reading you right, but does this mean you're focusing, then adjusting the defocus control. If so that will explain the front focus issues you are having as you need to focus after you've adjusted not before
Click to expand...

That would have been a problem Hugh but the Defocus Control was left in neutral throughout the whole process.

What "defocusing between each" means is that the after each test photo the lens was focused to infinity to ensure that it would have to actively refocus on the target each time.
 
You must log in or register to reply here.
Back
Top