Reading mode: Light Dark

New focus test

Go to last post
Thanks so much Tim for providing this to us!

Here is my result... looks pretty good to my eye, it'd be easier to tell with less DOF though.



Your opinions?
Tim Jackson said:
STOP! If you don't care about focus / back focus / front focus then
move right along to another thread.

To those of you still here, I have once again updated my notorious
focus test chart document.

There's now a LOT more information included on the AF system:

+ How the AF system works
+ Why the test chart works (in spite of rumours to the contrary)
+ Simple exercises for you to try out yourself which will greatly
help you in your understanding of how the AF system works.

Find it here: http://md.co.za/d70

Tim
--
=============================
At the time of posting this message, the latest
focus test chart is V2.1 (Dated 24 June 2004)
Get it here: http://md.co.za/d70
=============================
Click to expand...
 
Brad Berry said:
Here is my result... looks pretty good to my eye, it'd be easier to
tell with less DOF though.



Your opinions?
Tim Jackson said:
STOP! If you don't care about focus / back focus / front focus then
move right along to another thread.

To those of you still here, I have once again updated my notorious
focus test chart document.

There's now a LOT more information included on the AF system:

+ How the AF system works
+ Why the test chart works (in spite of rumours to the contrary)
+ Simple exercises for you to try out yourself which will greatly
help you in your understanding of how the AF system works.

Find it here: http://md.co.za/d70

Tim
--
=============================
At the time of posting this message, the latest
focus test chart is V2.1 (Dated 24 June 2004)
Get it here: http://md.co.za/d70
=============================
Click to expand...
Click to expand...
 
Hi msport,

I saw your post on this in another thread and have responded there so I won't duplicate all of it here.

Thanks for taking the trouble to go into detail. It all helps to get to the bottom of it.

As I mentioned in the other thread, I'd appreciate being able to see the full-size originals of your tests.

If you haven't already tried it, please would you do those same tests using my chart and let us know what your results are.

I've tried to do what you described (small changes to the vertical aim) with the camera in both portrait and landscape orientation but haven't yet been able to get my camera to do what yours was doing.

I was doing this using my chart which isn't identical to what you used so there may be an explanation in that, I don't know.

My findings have been that provided there is one single clear and unambiguous object for the AF system to lock onto then minor changes in where you aim don't affect the result.

If one was shooting a pic of a ruler, where there are lots of lines and numbers, I could easily understand why you'd get the result you were getting as the AF ssytem would have lots of stuff to focus on.

This is why I'd like to see your un-cropped originals, to try and gauge what the AF sensor would have been seeing.

Tim
--
=============================
At the time of posting this message, the latest
focus test chart is V2.1 (Dated 24 June 2004)
Get it here: http://md.co.za/d70
=============================
 
msport said:
I can repeatedly "make" my camera/kit lens front OR back OR dead-on
focus by making tiny adjustments to the vertical aim of my
camera.
Click to expand...
What do you mean by "vertical aim"? Changing the position of horizontal line you focus on like [-] and [ ]? Was it with D70?

--
no text
 
Iliah Borg said:
horizontal line you focus on like [-] and [ ]?
Click to expand...
Yes, that's what he meant.

As you look at the focus line on the chart, through the view finder, you would normally aim the camera so that the focus line passed through the centre of the focus area marker like this [ -- ].

Msport was experimenting with moving the focus area marker up or down a little so that the focus line was no longer exactly in the middle of the focus area marker but closer to the top or the bottom ([ _ ]) of the marker.

This would obviously move the focus sensor slightly up or down relative to the focus line on the test chart.

In my experience (using my chart) this hasn't made any difference to the test results but msport has had different results (using his chart).

I expect he'll give us more info when he next visits the forum.

Tim
--
=============================
At the time of posting this message, the latest
focus test chart is V2.1 (Dated 24 June 2004)
Get it here: http://md.co.za/d70
=============================
 
Tim Jackson said:
Iliah Borg said:
horizontal line you focus on like [-] and [ ]?
Click to expand...
Yes, that's what he meant.

As you look at the focus line on the chart, through the view
finder, you would normally aim the camera so that the focus line
passed through the centre of the focus area marker like this [ -- ].

Msport was experimenting with moving the focus area marker up or
down a little so that the focus line was no longer exactly in the
middle of the focus area marker but closer to the top or the bottom
([ _ ]) of the marker.

This would obviously move the focus sensor slightly up or down
relative to the focus line on the test chart.

In my experience (using my chart) this hasn't made any difference
to the test results but msport has had different results (using his
chart).
Click to expand...
I've tried this - moving the line to the top, to the bottom and to the middle of the bracket, using a good tripod with geared head. I have not seen any difference in focusing on 4 D70's I was experimenting with.

--
no text
 
Iliah Borg said:
I've tried this - moving the line to the top, to the bottom and to
the middle of the bracket, using a good tripod with geared head. I
have not seen any difference in focusing on 4 D70's I was
experimenting with.
Click to expand...
Iliah, thanks very much for posting this info.

My test results agree with yours.

When I've looked at the theory behind the AF system, I expected it to work this way and, when I've tried testing it in real life, as you have, I also got the same results, bearing out the theory.

I can only assume that in msport's tests there's some other factor that we aren't (yet) aware of.

My guess is that when msport did his tests, the AF sensor he was using was able to see more than just a single focus line.

Tim

--
=============================
At the time of posting this message, the latest
focus test chart is V2.1 (Dated 24 June 2004)
Get it here: http://md.co.za/d70
=============================
 
Tim Jackson said:
My test results agree with yours.
Click to expand...
As you probably remember, other CAM modules behaved different here in my tests.
Tim Jackson said:
When I've looked at the theory behind the AF system, I expected it
to work this way and, when I've tried testing it in real life, as
you have, I also got the same results, bearing out the theory.
Click to expand...
You know my reservations about your methodics :)

I never agreed with the target positioned non-parallel to the film plane, and never agreed that shooting this close is the way to test.

What is probably more important then my HO, cameramakers and repair shops also run the tests from greater distances and positioning focus targets in parallel with the sensor.

--
no text
 
Tailz said:
Iliah Borg said:
I never agreed with the target positioned non-parallel to the film
plane
Click to expand...
What do you mean, Iliah, in Tim's test the target is parallel to
the film plane. Or am I misunderstanding something ?
Click to expand...
I think there was a variant where the target was parallel; now in the description of the text (see Tim's signature for the link) the angle is 45 degrees, if I'm not missing something here.

--
no text
 
Tim Jackson said:
I can only assume that in msport's tests there's some other factor
that we aren't (yet) aware of.
Click to expand...
Might be because msport doesn't actually own a D70...

And why he'll never provide the full size images you requested, with exif data as I and others have requested many times.

spike
 
I must admit that my eyes are not as good as they used to be :((
So I've tried to quantify the BF tests using Tim's chart.

I captured the image per instructions.

The opened it in PS where I have Focus Magic as a filter. This filter works by analysing the out of focus pixels and shows a "Blur Width in Pixels" I dont pretend to know the algorhythmn used and I'm not sure it matters for this exercise. See the image at http://www.pbase.com/image/30566546 .

You can move the red box to calculate the blur width within it.

Now what I did was to start below the "ideal" focus line (ie closer to the lens) and "walk" the red box up the line and noted the numbers. The test image shown is from this thread. This is not ideal as the small size (and low PPI) makes the red box rather larger than I'd like. I have done this on my own full size images and the red box is about half the size.

As the image comes more into focus the the blurr width gets less. Again the actual number is not of concern to me - rather the trend of the numbers - do they go up or down.

Has anyone else tried this type of "objective" analysis?

David
Tim Jackson said:
STOP! If you don't care about focus / back focus / front focus then
move right along to another thread.

To those of you still here, I have once again updated my notorious
focus test chart document.

There's now a LOT more information included on the AF system:

+ How the AF system works
+ Why the test chart works (in spite of rumours to the contrary)
+ Simple exercises for you to try out yourself which will greatly
help you in your understanding of how the AF system works.

Find it here: http://md.co.za/d70

Tim
--
=============================
At the time of posting this message, the latest
focus test chart is V2.1 (Dated 24 June 2004)
Get it here: http://md.co.za/d70
=============================
Click to expand...
 
Hi,

Just gave the test a go, for some reason my paper seems to come our with a plue / purple tinge to it. I have tried changing the WB without any luck. Any ideas what would cause this?

Thanks
Davids8 said:
I captured the image per instructions.
The opened it in PS where I have Focus Magic as a filter. This
filter works by analysing the out of focus pixels and shows a "Blur
Width in Pixels" I dont pretend to know the algorhythmn used and
I'm not sure it matters for this exercise. See the image at
http://www.pbase.com/image/30566546 .

You can move the red box to calculate the blur width within it.

Now what I did was to start below the "ideal" focus line (ie closer
to the lens) and "walk" the red box up the line and noted the
numbers. The test image shown is from this thread. This is not
ideal as the small size (and low PPI) makes the red box rather
larger than I'd like. I have done this on my own full size images
and the red box is about half the size.
As the image comes more into focus the the blurr width gets less.
Again the actual number is not of concern to me - rather the trend
of the numbers - do they go up or down.

Has anyone else tried this type of "objective" analysis?

David
Tim Jackson said:
STOP! If you don't care about focus / back focus / front focus then
move right along to another thread.

To those of you still here, I have once again updated my notorious
focus test chart document.

There's now a LOT more information included on the AF system:

+ How the AF system works
+ Why the test chart works (in spite of rumours to the contrary)
+ Simple exercises for you to try out yourself which will greatly
help you in your understanding of how the AF system works.

Find it here: http://md.co.za/d70

Tim
--
=============================
At the time of posting this message, the latest
focus test chart is V2.1 (Dated 24 June 2004)
Get it here: http://md.co.za/d70
=============================
Click to expand...
Click to expand...
 
plue = blue!
cnewman1 said:
Just gave the test a go, for some reason my paper seems to come our
with a plue / purple tinge to it. I have tried changing the WB
without any luck. Any ideas what would cause this?

Thanks
Davids8 said:
I captured the image per instructions.
The opened it in PS where I have Focus Magic as a filter. This
filter works by analysing the out of focus pixels and shows a "Blur
Width in Pixels" I dont pretend to know the algorhythmn used and
I'm not sure it matters for this exercise. See the image at
http://www.pbase.com/image/30566546 .

You can move the red box to calculate the blur width within it.

Now what I did was to start below the "ideal" focus line (ie closer
to the lens) and "walk" the red box up the line and noted the
numbers. The test image shown is from this thread. This is not
ideal as the small size (and low PPI) makes the red box rather
larger than I'd like. I have done this on my own full size images
and the red box is about half the size.
As the image comes more into focus the the blurr width gets less.
Again the actual number is not of concern to me - rather the trend
of the numbers - do they go up or down.

Has anyone else tried this type of "objective" analysis?

David
Tim Jackson said:
STOP! If you don't care about focus / back focus / front focus then
move right along to another thread.

To those of you still here, I have once again updated my notorious
focus test chart document.

There's now a LOT more information included on the AF system:

+ How the AF system works
+ Why the test chart works (in spite of rumours to the contrary)
+ Simple exercises for you to try out yourself which will greatly
help you in your understanding of how the AF system works.

Find it here: http://md.co.za/d70

Tim
--
=============================
At the time of posting this message, the latest
focus test chart is V2.1 (Dated 24 June 2004)
Get it here: http://md.co.za/d70
=============================
Click to expand...
Click to expand...
Click to expand...
 
Tim,

Thank you much for the test. Below is what I get following your directions approx 98% of the time, although at different focal lengths and different lenses.

Sample of many many photos, this is representive of what I get, with my kit lens and my 28-300mm.
Tripod used, timer used.



I then did some real world experiments as well, and they also gave the same results as did your test. Please see photos below.I performed this test today in direct sunlight, the first in many days : )

Overview (top view) of simple setup as this was not needed to measure the degree of backfocus, I used it simply to confirm previous testing, in better lighting.



The results verified Tims test. See below



I would like to thank Tim for all the work he has put into this, and he should be commended for his efforts, from those that benifit from the work he has done, and those that do not. Although you may not approve of the methods, and or disagree with them. The fact of the matter is, he is attempting, and succeding at doing something... Informing people that there is a problem, and providing a means with which you are able to trest at home. I would not have tested had I not felt there was something wrong. Regardless if something is wrong or not, the D70 is under warranty, and costs me nothing to send it back for tests, and if need be, a tune up :)

Best regards,
Thomas

Keep up the good work Tim!
Tim Jackson said:
STOP! If you don't care about focus / back focus / front focus then
move right along to another thread.

To those of you still here, I have once again updated my notorious
focus test chart document.

There's now a LOT more information included on the AF system:

+ How the AF system works
+ Why the test chart works (in spite of rumours to the contrary)
+ Simple exercises for you to try out yourself which will greatly
help you in your understanding of how the AF system works.

Find it here: http://md.co.za/d70

Tim
--
=============================
At the time of posting this message, the latest
focus test chart is V2.1 (Dated 24 June 2004)
Get it here: http://md.co.za/d70
=============================
Click to expand...
--
http://www.280dgamingclan.com/ThomasAdams/gallery/
 
Great Job Tim.
Tim Jackson said:
Hi msport,

I saw your post on this in another thread and have responded there
so I won't duplicate all of it here.

Thanks for taking the trouble to go into detail. It all helps to
get to the bottom of it.

As I mentioned in the other thread, I'd appreciate being able to
see the full-size originals of your tests.

If you haven't already tried it, please would you do those same
tests using my chart and let us know what your results are.

I've tried to do what you described (small changes to the vertical
aim) with the camera in both portrait and landscape orientation but
haven't yet been able to get my camera to do what yours was doing.

I was doing this using my chart which isn't identical to what you
used so there may be an explanation in that, I don't know.

My findings have been that provided there is one single clear and
unambiguous object for the AF system to lock onto then minor
changes in where you aim don't affect the result.

If one was shooting a pic of a ruler, where there are lots of lines
and numbers, I could easily understand why you'd get the result you
were getting as the AF ssytem would have lots of stuff to focus on.

This is why I'd like to see your un-cropped originals, to try and
gauge what the AF sensor would have been seeing.

Tim
--
=============================
At the time of posting this message, the latest
focus test chart is V2.1 (Dated 24 June 2004)
Get it here: http://md.co.za/d70
=============================
Click to expand...
 
I am an owner of the Nikon N80, and D70, please see above for my photos of this issue. I also used an alternative tes, a real world test if you will, where out of 20 shots, I recieved the same result. I am not blaming anyone for this, I am simply aware of it, and when I am able to I will send it in for adjustment.The 4th of July is coming up, as is the birth of my 3rd child.

Best regards,
Thomas Adams
Mikhail Bogdanov said:
THomasAdams said:
I believe he is supplying a way to test our excellent cameras for
an issue that can be fixed easily, so that we can continue to enjoy
our D70's!
Click to expand...
Click to expand...
--
http://www.280dgamingclan.com/ThomasAdams/gallery/
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top