Lens Fine Tune Overrated?
I just got around to playing with the lens fine tune function on the D300. I have the Sigma 30 1.4, Nikkor 80-200 2.8 two ring, and the Tamron 17-50 2.8. I placed my camera on a sturdy bogen tripod and shot different assortments of books and curved medicine bottles to try to determine if any of my lenses could be more accurate after fine tuning.
What I found was all of my lenses looked best without any tuning at all and each one at +5 showed very little difference from 0 when zoomed in on the camera's LCD. I shot all these lenses wide open and at the furthest focal lengths. In the case of the rounded medicine bottles, I found that overall image sharpening probably would make more of a difference then slight fine tuning on the camera.
Has anyone else had any luck with this feature and is there any reason to think all our lenses suffer from front/back focus? If you read these boards it seems like every lens is plagued but I feel pretty good that my lenses are within reason for focus. I know that if I were to print 4x6s or 8x10s, the focus is definitely spot on where you couldn't notice a couple mm off.
Maybe I need another method for testing or maybe I'm looking for differences among too small of a distance (the distance between the text on the edge of a multi vitamen bottle and the part closest to the camera is probably half an inch).
The built-in LCD is WAY too small to make any judgement about how your pictures will look with AF fine tuning. Use a huge monitor and view at 50 percent. It takes a LOT of work to get the AF fine tuning right, and most of time you WON'T need it, but if you have a hunch that you do, then probably you do. The best method is to use Camera Control Pro 2 and see all your test shots in near real time. First start by making a picture with the AF fine tuning off, then start at +5, then -5 and objectively compare them to the original. If you see the picture getting worse or losing tack sharpness on the focal point, then just leave AF fine tuning off for that lens. If you see a slight improvement, play with it and adjust to convergence between your two starting points.
While many, or even most, may not need it, it does work for the people that do have issues. As was mentioned, use a large monitor and also look at how others have checked out their focus
I just got around to playing with the lens fine tune function on the
D300. I have the Sigma 30 1.4, Nikkor 80-200 2.8 two ring, and the
Tamron 17-50 2.8. I placed my camera on a sturdy bogen tripod and
shot different assortments of books and curved medicine bottles to
try to determine if any of my lenses could be more accurate after
What I found was all of my lenses looked best without any tuning at
all and each one at +5 showed very little difference from 0 when
zoomed in on the camera's LCD. I shot all these lenses wide open and
at the furthest focal lengths. In the case of the rounded medicine
bottles, I found that overall image sharpening probably would make
more of a difference then slight fine tuning on the camera.
Has anyone else had any luck with this feature and is there any
reason to think all our lenses suffer from front/back focus? If you
read these boards it seems like every lens is plagued but I feel
pretty good that my lenses are within reason for focus. I know that
if I were to print 4x6s or 8x10s, the focus is definitely spot on
where you couldn't notice a couple mm off.
Maybe I need another method for testing or maybe I'm looking for
differences among too small of a distance (the distance between the
text on the edge of a multi vitamen bottle and the part closest to
the camera is probably half an inch).
Okay....here's the deal....
1) Not every lens is going to need adjustment. While some of my lenses needed it, others didn't. Your lenses may not need any or very little adjustment. Your ext lens purchase may be different.
2) The rear LCD is just fine for making adjustments. Perfect in fact.
3) Let's not call a feature overrated just because "some" people don't need it. The D300 feature set is huge.
PS: There was some info about 3rd party lenses not being remembered by the D300 because the camera could not differentiate between them. Look into it.
Thomas (Lord Nikon!)
I don't consider it overrated at all, if your lens actually needs it (untuned vs +15, full crop at 300mm)
I did the comparison by taking shots at -20 to +20 in steps of 5, inspecting the results on my computer, and isolating from there.
my little experience
Civilization was not born from labour, but from freetime and game
only if in focus picturs are over rated...
I have a Sigma 70-200 that needs +19 to get it right....so for me....its not over rated.
Just consider yourself lucky that your lenses dont need much tuning.
'Miles to go before I sleep.'
after loading my test shots into photoshop on my computer, I'm having a hard time even telling a difference between 0, +5, and +10. I layer each image over each other and then click them on and off to see if I can tell which one looks sharper. At 100% it's really hard to tell so I'm thinking when the images are printed or resized for web you won't notice at all. Should there be a huge difference between 0 and +10?
I think the issue is a complicated one. Like, not all adjustment values work for all degrees of focal length or distance.
I sat down and did a test with my 50mm 1.8 and saw a minute difference. So I went through my 3 lenses and tested and set the AF tune for each one.
Then I took a picture of my son, and though it looked a little soft (pixel peeping) so I turned off the AF fine tune and that image was noticeably sharper.
I may re-investigate it as I am something of a sharpness freak. But for now I've got it off.
Everything I write is a personal opinion. Even when I quote facts, they are the facts I personally choose to accept.
Yeah I'm noticing the same thing. I run some test and 'think' some setting of fine tuning looks best. Then I reshoot some stuff around the house and start thinking maybe it's soft again. Then I get paranoid that maybe my zooms are WAY off at a focal length I didn't calibrate for. Overall I'm thinking I didn't notice any problems with my D200 so maybe I shouldn't mess with it. It seems like a good function but like someone else said above, people have been relying on AF now for 20 years or so with no major problems.
I honestly think a lot of people on these boards harp on the front/back focus thing too much. I work at a camera store and have people coming in all the time who act like they know what they are talking about and throw the back/front focus argument around a lot. I just don't get it but since most of my shooting is at 2.8 or faster, I'd like to know I have achieved the maximum sharpness from my camera/lens.
When I try to run my test on lettering such as a book that is tilted 45 degrees away from the lens plane, I notice a lot of CA which makes it really hard to tell if the focus is perfect. I guess if the lens is focusing more or less what you are focusing on then it should be fine but I think this whole "fine tuning" has me believing it should be perfect down to a centimenter or so. Maybe this feature really is for extremely far off focusing problems on certain lenses. Welp I'm assisting with a wedding tomorrow so I'll soon see how the D300 fairs for the first time.
...I have a Tamron zoom that has been fine for me...I downloaded a focus test chart from one of these forums...and shot it at a 45 degree angle...went into the menu after looking at the shot on the LCD...did a -1 click in AF fine tune...the camera recognized the lens...I hit "enter"....and I shot some more both at close/far distances, and it seems at this point that I had "sent the lens in for calibration"...it is sharp as a razor to me....a little sharper than it was (at least this is my impression)...so I'm a happy camper at this point. I am a 30 year shooter and not an amateur so I have noticed lens sharpness for years, and I think small tweaks with the AF fine tune can actually be of benefit...I don't know if I would go to the "15 or 20 clicks" route..but this one click seems to have optimized this Tamron (28-75).
This would be what you'd do if a zoom was parafocal --- focus doesn't change as you zoom --- but, nearly all modern zooms ARE NOT parafocal.
The older manual focus zooms were, but that's not relavent here . . .
This should be good for primes and parafocal zooms.
For current zooms, shouldn't you test at the middle of the focal length range??? Point of best focus varies with focal length for these -- that's why their users manuals all say zoom and then focus . . .
[Equipment in use -- see profile]
good question Andy. I just assumed you'd test it at the furthest focal distance because don't images get sharper the more wide angle you go? DOF becomes more shallow at the telephoto end so I thought this might become the most crucial focal length of the zoom lens to test the calibration on.
And what is this feature actually doing? Is it physically moving the sensor or something or is it some sort of software calibration within the lens? Must be the lens because it seems ridiculous to think it's the sensors.
That's what you'd do for a parafocal zoom, but not the newer AF/AFS zooms. When I shot MF, the zooms were parafocal. I used to zoom to max. focal length, focus, and then adjust zoom to lessor focal length to frame my shot, and then shoot away. You metered wide open & focused on a ground glas screen. So, always razor sharp.
Can't do that with the newer, vari-focal, AF/AFS zooms. Have to frame, focus (manually or auto), and take shot. You still focus wide open (because that's where you are for metering) AF or MF, but you have to do it at your framing focal length. When you actually take your shot, the lens stops down, which is why it's important to do test shots at wide open aperture.
Testing at a middle focal length seems to make sense, but it presumes any focusing error is constant as you zoom in or out from this point. Which isn't necessarily the case -- would be surprised if the zooms were actually designed this way.
When you send one off to be calibrated, they take readings at multiple focal lengths and then, I presume, do a treak that gives best, on balance, overall performance. Makes this "simple" adjustment not so simple. Because for best results, you'd also have to check you zoom at multiple focal lengths. And choose a compromise correction. This is lens calibration.
Here, Nikon is ducking lens calibration and having you calibrate the camera for each lens. Saves them the hassle and expense of calibrating lenses that are off. Nice, huh?? There's no free lunch.
The few times I've checked one of my zoom lenses because of some reputed horrible sample variation, I didn't get the exact same focal point at every focal length. But as long as it came out within the in-focus range, in front or back of the aiming point within Depth-of-Field for max. aperture, I considered it to be OK. And they were always OK -- but not bang-on at every focal length -- just OK.
[Equipment in use -- see profile]
If you do a search on "backfocus" here, and read enough posts, you will find a small number if shooters that reported that they sent a body and multiple lenses to Nikon for calibration, usually because one single lens was not focusing correctly. They then reported that Nikon told them that they could have the camera calibrated for the one lens, or the other lenses, but Nikon could not get all the lenses to focus perfectly. Obviously these shooters, with, say, a D2X, 70-200VR, 17-55DX and a couple other expensive lenses, weren't too pleased and there was much discussion here about the idea of in-camera lens specific user calibration.
(I thought... yea right... we'll see that soon...NOT... but I guess I was pleasantly wrong)
Assuming the reports were accurate, this is a HUGE thing for what is probably a small minority of users, plus others with real or imagined greater powers of sight and more precise requirements than Nikon delivers
If I had single lens focusing badly enough to be actually reflected in my photos, I'd have Nikon calibrate the one lens.
I have twelve -- not cheap -- lenses. Only one of them is off a bit, but well within DOF at max. aperture.
So, practically speaking, its being slightly off has no consequence in terms of REAL image sharpness. Not worth sending to Nikon, who would just send it back telling me it was in-spec.
Defective lenses with calibration problems should IMHO be fixed by the lens maker.
But, yes, there are folks with
front or backfocus that having this feature will be HUGE to. And will rush out to buy the D300 to correct it. Good marketing for Nikon.
Don't get me wrong, it's a nice feature, but really mostly unnecessary.
[Equipment in use -- see profile]
A couple of weeks ago, someone posted a series of photos showing AF fine tuning test results, and they showed a clear difference, even between 0 and +5 settings. The shots were of a line of soda cans, with the line at an angle to the photographer, and focussed on the fine print on the center can. Pretty convincing results.
I Reject Your Reality And Substitute My Own
Web Site - http://www.hgiersberg.com/
|National Gallery of Art by Kukla|
from Your City - Black and White (in colour!)
|Hummingbird and Bee by dibilio57|
from My Best Photo of the Week
|The Snowy Egret by Lee8282|
from Color - Monochrome
|Skate Boarder dpr-0927 by vbuhay|
from Skateboarding Cover shot