Again with the focusing

Joined
May 5, 2005
Messages
482
Location
Rexburg, ID
I'm sure you guys are sick of hearing from me but untill someone tells me to shut up, I'll keep posting :). After looking at my situation, I wanted to see if my camera has focus with all AF-S lens or just the couple that I own. I went to another store and tried a 17-55, 70-200 VR, and 17-35. All 3 of these lens performed the same...center focus perfect every time, other sensors hit and miss. They also had a D2x body so I tried it as well. It was the same as mine.

This now means I've tested 4 D2x bodies, 2 Nikkor 12-24 lens, a Nikkor 70-200, a Sigma 70-200, Nikkor 17-55 & 17-35, a Sigma 105, a Nikkor 50 1.4, a Tamron 28-75, and a Tokina 12-24. My conclusion so far is that the D2x focus better with screw type lens.

I guess the next step is to hunt down a D2h or D2hs and compare them.
 
Joined
Jul 4, 2005
Messages
94
Ruffles

How do you test a lens so fast, it took me all day to test 3 lenses?.

Keep on blabbing. :lol:

Butch
 
Joined
May 5, 2005
Messages
482
Location
Rexburg, ID
Sorry it took so long to reply. I've been kicking butt in the Sportsman class with a TQ and Win at round 6 of the NATC at Timezone.

As far as how I test my lens, it's pretty simple and I'm open to the possibility that it's flawed but I'll go into that more later. What I do is this...

Camera Setup
Turn image review to on.
Set the center button in playback mode to toggle the zoom (set to medium)
Autofocus mode set to AF-S
Autofocus area mode set to Single Area AF

Testing
Select focus sensor
Turn lens focus ring to infinity
Focus on subject (verify green focus confirmation dot)
take picture
press center button to zoom in on selected focus sensor
See if it's sharp
repeat for each sensor

While some people might say that you can't judge sharpness on the camera LCD and that you need to apply USM, my experience has been that you can easily tell the difference between shots done with the center sensor and others.


I talked to Nikon about my problems and got some interesting information. First, the way they test focus is very similar to the paper target thing but with one very big difference. The tech said that the focus target needs to be parallel to the CCD. The target they use has the distance scale at a 45 degree angle but at 0, they have a big flat panel that sticks up. They focus on the panel and after the camera locks on it, remove the panel so they can see the entire scale. Then they can see if the focus is in front or behind the 0 point.

I decided to try something similar. In my previous tests, I've always focused on the SAME target with each sensor. As the selected sensor gets farther away from the center, you have to turn the camera more to put the sensor on the subject. This put the target further and further away from parallel to the CCD. I tried setting my camera on a tripod facing directly at my computer monitor (flat panel parallel to the cameras CCD) and while not moving the camera at all, taking a picture of the monitor with each sensor. I changed the lens to infinity between each shot to ensure that the camera had to re-acquire focus. The result was perfect focus with each sensor on every lens.

Now that I'm able to reproduce Nikon's results, I need to think about how this applies to typical shooting situations. I explained my testing procedure to Nikon and while the tech pointed out the parallel issue, he said I should still see good results. He was not opposed to pursuing a replacement camera but as I've had the same issue with 4 D2x's, neither of us was sure that was the right thing to do. Perhaps I just need to understand the limits of the camera and learn to work within them.

On a separate note, I spent the weekend shooting the Tamron & Tokina lens I picked up and was able to reproduce the same thing so I guess it's not a screw drive lens thing after all. The Tamron was very soft @ F/2.8 and the Tokina had LOTS of CA...much worse than the Nikkor 12-24 so I returned both of those lens today. I've got a busy week at work so I'll have to wait until next week to test some more and let you know what I find and if I figure out a better technique.
 
Joined
Jul 4, 2005
Messages
94
"The tech said that the focus target needs to be parallel to the CCD".

Very interesting, that's why a lot of people was getting bad results.

I think it would be easy to make a panel like the one Nikon uses. I have no reason to check mine anymore,it's dead on.

Don't forget to check the wide end of the zooms,it's easier to check.

Butch
 
C

caero

Guest
The 45 degree test chart method has been proven a bit faulty, as the AF sensors have a very hard time picking up that little black line in that angle.

Instead try one of the following.

Line up 3 colourful soda cans a bit like this:
Code:
O
 O
  O
With them overlapping each other marginally depth wise.

Then place your tripod and camera directly in front of the middle center can. Focus on that one and take a few exposures.

Remember to have good lighting (maybe even do it outside in semi cloudy weather on a garden table) so that dull lighting can't be blamed.

I know that you won't get mm precision measures, but you should be able to see which can is most in focus and how the dof is adjusted (front or back biased).

I've also seen another guy mentioning a test using a big field of flowers. But I think that is a bit overkill to drive around looking for one :)
 
Joined
Apr 30, 2005
Messages
7,818
Location
Gilroy, California
What I have used is a bit of text, more or less parallel to the sensor, surrounded by half tone printing since it shows sharpness so well.

Then set that on a surface with a fine pattern to it like window screen, etc. at an angle so I can see where the DOF lands.

I don't see why the line in that popular chart isn't a decent target though. The fact that it is at an angle is irrelevant to the AF system. You can fool yourself so far as where the sensor actually is and how square the camera is, but otherwise the few times I have tried that chart it seemed reliable.

But mostly, I just look at the results of real photographs.
 

Latest posts

Links on this page may be to our affiliates. Sales through affiliate links may benefit this site.
Nikon Cafe is a fan site and not associated with Nikon Corporation.
Forum post reactions by Twemoji: https://github.com/twitter/twemoji
Copyright © 2005-2019 Amin Forums, LLC
Top Bottom