d7000 AF Tune result

[pekklemon]

hi guys, i've recently acquired a d7000 and am enjoying every minute of it. I just did some focus testing out of boredom for my nikkor 35mm f1/8G lens.

The result is a little perplexing for me. It seems like its quite on point in the middle but i notice on the left side of the chart, it seems to be back focusing but on the right side of the chart, it is front focusing slightly.

Subscribe to see EXIF info for this image (if available)

is this normal?

 

[sylvesterii]

Did you do this hand-held? If so, you may not have been perfectly parallel with the chart. Thus the "line" of in-focus is running right through the middle but at a slight angle. Hence a little backwards on the left, a little forwards on the right.

 

[trenchmonkey]

+1 Bored, take pictures. AF fine tune in the wrong hands can muck up perfectly fine glass.

 

[Holmes]

+1 Bored, take pictures. AF fine tune in the wrong hands can muck up perfectly fine glass.

Agreed.... and speaking from personal experience :biggrin:

 

[Growltiger]

Never base fine tune on focussing on a test chart held at an angle. The focus can easily be thrown out and you will calibrate all the lenses wrongly.

A simple and better method is to take a series of photos of a good flat-on target, changing the AF Fine Tune to all the likely settings one after another. Then compare them on your computer. Slow but more reliable.

 

[Patrick]

Never base fine tune on focussing on a test chart held at an angle. The focus can easily be thrown out and you will calibrate all the lenses wrongly.

A simple and better method is to take a series of photos of a good flat-on target, changing the AF Fine Tune to all the likely settings one after another. Then compare them on your computer. Slow but more reliable.

Big steps can be done without a big screen . the real Finetune-part should be done with a computer yes, I've used tethered mode for that !

 

[pekklemon]

Did you do this hand-held? If so, you may not have been perfectly parallel with the chart. Thus the "line" of in-focus is running right through the middle but at a slight angle. Hence a little backwards on the left, a little forwards on the right.

i did do this particular shot handheld. I will do one with a tripod tonight. I havent really come across any issue with AF with the 35mm lens. I should do as trenchmonkey suggested and just take photos when im bored ><

 

[wizzard0003]

+1 on never set AF fine tune by focusing on an angled target...!

For reference, check out how the "Lens Align" system works...

http://www.whibalhost.com/lensalign/

You'll notice there is a flat target that you focus on and the scale is to the side... You set
up your camera on a tripod and make sure it's square and perpendicular to the flat target...

You should also use mirror up and a timed delay to eliminate any blurr caused by
mirror slap vibration, etc... :wink:

 

[photoforfun]

Never base fine tune on focussing on a test chart held at an angle. The focus can easily be thrown out and you will calibrate all the lenses wrongly.

A simple and better method is to take a series of photos of a good flat-on target, changing the AF Fine Tune to all the likely settings one after another. Then compare them on your computer. Slow but more reliable.

+1. Exactly my method. I do all shots twice from -20 over 0 to +20 and compare them in capture NX2 side by side.
Made my target myself and the results are as what I expect.
For focusing distance I take bewteen 10-15 times focal or biggest focal if it is a zoom. For macro lens I test very close(that's where macro lenses are made for I guess... )
Here my experience with my severe backfocusing D7000 with kitlens and 105VR. http://www.fotografie.fr/fotoforum/viewtopic.php?f=54&t=1140

Here the target I use:

Subscribe to see EXIF info for this image (if available)

 

[bh626]

Because of soft pictures, I tested the kit lens for focus accuracy. I think there may be an issue with it as I need -8 fine tune and I have read this several times as well. On the other hand the 70-300 and 35 seem very sharp in real photos - probably why I haven't tested them.

Here is a thread about a guy's experience with the camera sent into Nikon for backfocus issues, after several trips he still needed -8 for his kit lens - http://www.digitalhome.ca/forum/showthread.php?p=1188260#post1188260

Here's my 35mm, 0 fine tune

Subscribe to see EXIF info for this image (if available)

and 70-300, 0 fine tune

Subscribe to see EXIF info for this image (if available)

 

[pekklemon]

Never base fine tune on focussing on a test chart held at an angle. The focus can easily be thrown out and you will calibrate all the lenses wrongly.

A simple and better method is to take a series of photos of a good flat-on target, changing the AF Fine Tune to all the likely settings one after another. Then compare them on your computer. Slow but more reliable.

could you please do a step by step instruction on what i should do to test out the focus properly? Is the chart that I am using sufficient? I was following Jeffrey Friedl's auto focus instructions but from what I hear it is not the best method to test AF.

Correct me if I'm wrong, so what I need to do is instead taking an angled shot, I should take a flat shot such as putting the chart up on the wall and use a tripod to take the shot. How far away should I take the shot from?

Im quite new to this so I very much appreciate your guidance

 

[Holmes]

This might be of interest to you:

http://blog.nikonians.org/archives/2008/03/id61_the_image_1.html

 

[Growltiger]

could you please do a step by step instruction on what i should do to test out the focus properly? Is the chart that I am using sufficient? I was following Jeffrey Friedl's auto focus instructions but from what I hear it is not the best method to test AF.

Correct me if I'm wrong, so what I need to do is instead taking an angled shot, I should take a flat shot such as putting the chart up on the wall and use a tripod to take the shot. How far away should I take the shot from?

Im quite new to this so I very much appreciate your guidance

Basic setup
Stick the test chart on the wall on the far side of the room, not close-up.
Stand on the far side of the room and put the camera on a tripod aimed at the chart, straight on, not at an angle.
Set to A exposure mode and set the aperture wide open.
Set focus to single point, and set the focus point to the central position.
Have enough light for reliable focus.

Procedure
Now set the AF Fine Tune to -15 and take a photo.
Change it to -14 and take a photo.
Keep changing it by 1 at a time.
Continue until you get to +15.

Now copy all the 31 photos to your computer and study them.

This will give you a far more reliable result.

 

[bh626]

Here's a tip to make sure the camera is straight on. Tape a mirror flat on the target. When looking through the VF you will be straight vertically and horizontally when you see the lens in the center of the mirror.

Nikon service uses a rig whereby the camera is on a track perpendicular to a flat target. There is a ruler connected to the target at a 45 degree angle, with the 0 mark being even with the front of the target - kinda like the lens align tool. This is the most accurate way to ensure reliable testing since you are focusing on a flat target that is large enough to compensate for the fact that the focal point maybe not be exactly where the VF says it is.

You could make your own test target pretty easily to simulate this - take a look here http://www.peleng8.com/how-to-detect-back-focus.html

 

[Takeda]

+1 on never set AF fine tune by focusing on an angled target...!

For reference, check out how the "Lens Align" system works...

http://www.whibalhost.com/lensalign/

You'll notice there is a flat target that you focus on and the scale is to the side... You set
up your camera on a tripod and make sure it's square and perpendicular to the flat target...

You should also use mirror up and a timed delay to eliminate any blurr caused by
mirror slap vibration, etc... :wink:

+1 Over and over people are told not to do AF fine tune with an angled target, that the target should be parallel to the camera / image sensor.

And on top of that, they try to do an AF fine tune WIDE OPEN, on a lens that is soft wide open!

Simply amazing!

 

[sylvesterii]

to be fair, it is probably a good idea to test wide open and stopped down. Some lenses exhibit focus shift (some significantly), and a lot of people who buy fast glass buy it to shoot wide open.

 

[pekklemon]

Basic setup
Stick the test chart on the wall on the far side of the room, not close-up.
Stand on the far side of the room and put the camera on a tripod aimed at the chart, straight on, not at an angle.
Set to A exposure mode and set the aperture wide open.
Set focus to single point, and set the focus point to the central position.
Have enough light for reliable focus.

Procedure
Now set the AF Fine Tune to -15 and take a photo.
Change it to -14 and take a photo.
Keep changing it by 1 at a time.
Continue until you get to +15.

Now copy all the 31 photos to your computer and study them.

This will give you a far more reliable result.

Thanks, I'll give this a shot!

Just another question while im at it, would you guys consider these images 'tack' sharp?

Tamron 17-50
http://www.flickr.com/photos/scribblenauts/5657794996/in/photostream

Nikkor 35mm 1/8
http://www.flickr.com/photos/scribblenauts/5674819763/in/photostream
http://www.flickr.com/photos/scribblenauts/5675380256/in/photostream

Nikkor 50 1/4G
http://www.flickr.com/photos/scribblenauts/5624842530/in/photostream

 

[Growltiger]

Thanks, I'll give this a shot!

Just another question while im at it, would you guys consider these images 'tack' sharp?

Tamron 17-50
http://www.flickr.com/photos/scribblenauts/5657794996/in/photostream

Nikkor 35mm 1/8
http://www.flickr.com/photos/scribblenauts/5674819763/in/photostream
http://www.flickr.com/photos/scribblenauts/5675380256/in/photostream

Nikkor 50 1/4G
http://www.flickr.com/photos/scribblenauts/5624842530/in/photostream

You cannot possibly tell if they are sharp.

You need to publish 100% crops, then we can all pixel-peep.

This means you crop an area no more than 1024 pixels wide and please link to it so we can see it here, not on another website. You can put Flickr images here. (I'm assuming that Flickr themselves don't degrade the images - I know some photo sharing websites do that.)

 

[bh626]

Thanks, I'll give this a shot!

Just another question while im at it, would you guys consider these images 'tack' sharp?

Pretty sharp, but light plays a very important role in how sharp a picture looks, and the lighting in those is not that good