AF micro-adjust made easy

I think the 7D ML is roughly where the 5D3 one is - in alpha testing so they're being a bit cautious in case it breaks anything. That said for this use it shouldn't matter - just load, test and off - no more ML until you want it again.

Also I don;t think there a pre-prepared bin file yet, so if you wanted to go that route you'd have to compile it yourself (something I would do if I knew how)

I have it on my camera but almost never load it currently, but would be happy if it were on all the time as I used it back on the 60D without any problems at all.
 
Do you mind me asking how far away you put you target?

I'm curious as Canon say range x50, others say x20
 
JohnN said:
Do you mind me asking how far away you put you target?

I'm curious as Canon say range x50, others say x20
Click to expand...

I was useing the x50 route

It was ok on my shorter lens but ran out of room for the longer lens, I will have to do this outside.
 
I remember trying to do a 500mm at x50, eventually I gave up and sent it to Sigma :)
 
There is no set distance for testing, other than it must not be too close or you might end up with settings that are out at longer range. And it can't be too distant or there'll be too much depth of field to see clearly what's happening.

Don't know if this is helpful, but almost any lens longer than a wide-angle* should be capable of performing well within it's optimum range for a decent head and shoulders portrait. Not a bad starting point.

*Wide angles are difficult. Often used for distant landscape views but usually the only way you can see what's happening with focus is to test them very close to the target. Frankly I don't bother much with wide-angles. I just check that it's not miles out, basically that it's not faulty, and leave it. I'd take more care with something like a 35mm f/1.4, and do the test at the furthest distance where I could still reliably see the exact focus position on the target/ruler/cerial box I posted earlier. A larger target and longer rule helps when DoF is generous.
 
Nice! just done two bodies, four lenses and extender/lenses combos - took four hours lol, but hopefully results will be worth it (ended up in the garden :lol:)

only one that didn't work was 2x on the 300/2.8 on the 1d4 @ 39metres. couldn't get a reading +/- 20 even when de-focusing, rest were good

wish 1d4 had the same AF menu as the 5D3, i.e. programming in a wide and telephoto value rather than one for the whole range
 
Last edited:
Although i did buy FoCal last month ive yet to use it.
Ive been using the moire pattern method and getting good results (as in no MFA needed), and it doesn't require any test charts printed out, but i am going to give the latest FoCal a go this week, just to see if it makes any difference or tells me something i wasn't aware of.
http://www.northlight-images.co.uk/article_pages/cameras/1ds3_af_micoadjustment.html
 
Last edited:
Focal 1.8 was released today, so make sure you're up to date.
 
tdodd said:
Focal 1.8 was released today, so make sure you're up to date.
Click to expand...

Cheers. yes, ive already got it. When i first brought it i couldn't get v1.7 to work, it kept crashing but v1.8 RC seemed stable, although i never got around to actually running any tests.
 
So, I thought I'd check my 24-70mm f/2.8 lens today and got the following:
@ 70mm : +2 to -11 so -4 mid-point
@ 50mm : +6 to -9 so -3 mid-point
@ 24mm : +10 to -4 so +3 mid-point

Leaves me a little confused about what setting to use :shrug:
 
Doesn't the D300 allow profiles for several lenses?
 
Bristolian said:
So, I thought I'd check my 24-70mm f/2.8 lens today and got the following:
@ 70mm : +2 to -11 so -4 mid-point
@ 50mm : +6 to -9 so -3 mid-point
@ 24mm : +10 to -4 so +3 mid-point

Leaves me a little confused about what setting to use :shrug:
Click to expand...

Some cameras, like the Canon 5D Mk3, allow you to apply different settings for the long and short ends of the zoom range.

I also saw a demo of Sigma's new USB Dock thingy at Focus show today, this jobbie http://www.techradar.com/news/photo...as/sigma-set-to-enable-af-adjustments-1100933 For the price of a good filter, it allows you to set AF microadjustments for four different focal lengths of the zoom range, and also four different focusing distances which is a cool option I've not seen before.

It also allows you to do other things, like upgrade lens firmware and fiddle with image stabilisation (mainly for smoother video I think) but only works on certain Sigma lenses. On the other hand, I'm beginning to wonder if we can have too much of a good thing?
 
Would the focus be less critical at 24mm?
If so bias towards the other end if anything.
 
HoppyUK said:
Some cameras, like the Canon 5D Mk3, allow you to apply different settings for the long and short ends of the zoom range.
Click to expand...

Unfortunately I don't think Nikon do that ... certainly the D300 doesn't.

I also saw a demo of Sigma's new USB Dock thingy at Focus show today, this jobbie http://www.techradar.com/news/photo...as/sigma-set-to-enable-af-adjustments-1100933 For the price of a good filter, it allows you to set AF microadjustments for four different focal lengths of the zoom range, and also four different focusing distances which is a cool option I've not seen before.

It also allows you to do other things, like upgrade lens firmware and fiddle with image stabilisation (mainly for smoother video I think) but only works on certain Sigma lenses. On the other hand, I'm beginning to wonder if we can have too much of a good thing?
Click to expand...

That's certainly something I'll be keeping an eye on. Thanks for the link.
 
4wd said:
Would the focus be less critical at 24mm?
If so bias towards the other end if anything.
Click to expand...

Yes, that's what I'm thinking. To be honest, I can't physically see the difference between two identical images (taken during my tests), one at zero AF adjustment and the other at -4 so I'm wondering if it's worth the bother at all.

Might be interesting to try my Sigma 150-500 though and see what results I get :D
 
Bristolian said:
To be honest, I can't physically see the difference between two identical images (taken during my tests), one at zero AF adjustment and the other at -4
Click to expand...

And I don't think you're alone there. You're just unusual in pointing it out.

I keep saying this - it's called micro focus adjustment for a reason. People get all het up about differences of 1 or 2. But if you use something like FoCal Pro and examine the plots it produces then you'll see that, in most cases, you could pick anything over a range of +/- 4 units and the results would be within the range of experimental error - they would give indistinguishable real-world results.
 
hollis_f said:
And I don't think you're alone there. You're just unusual in pointing it out.

I keep saying this - it's called micro focus adjustment for a reason. People get all het up about differences of 1 or 2. But if you use something like FoCal Pro and examine the plots it produces then you'll see that, in most cases, you could pick anything over a range of +/- 4 units and the results would be within the range of experimental error - they would give indistinguishable real-world results.
Click to expand...

Very true. This is just another form of pixel peeping and we all know that usually ends in tears LOL

Half of me says, I want this thing to be as accurate as it possibly can be, and the other half of me knows that in reality it rarely matters, and when it does, there's a whole raft of other technique issues that render absolute accuracy impossible anyway.

At the end of the day, what matters to me is that my kit is performing to its design spec, basically that there is no serious mis-match error between the lens and camera calibration. With that reassurance, if it's -2 at one end of the zoom and +2 at the other, I'm not going to worry. Given a larger difference, if I had to choose, I'd err towards bias at the longer end where depth of field is more likely to be shallow.
 
The you tube video has an interesting update to to description. Essentially it says when a Nikon is set to MF the in focus range I bigger so he recommends you use back button focus instead. This should work on even cheaper nikons where you can set the AE/AF lock to AF on to do the tune.
 
Bristolian said:
Yes, that's what I'm thinking. To be honest, I can't physically see the difference between two identical images (taken during my tests), one at zero AF adjustment and the other at -4 so I'm wondering if it's worth the bother at all.

Might be interesting to try my Sigma 150-500 though and see what results I get :D
Click to expand...

I feel the same. It amazes me that people buy a new lens and immediately have to do some MFA to it, just to feel better in most cases.
I can understand if the lens is faulty or soft, but i dont believe that using MFA on a good lens makes it a better lens, at least not for every image.
 
Well I tried this today and wasn't able to get any of my lenses in the +20/-20 range!
Maybe I'm doing something wrong!
They all lost focus in the +range but still had a constant focus confirmation dot at -20.
I am aware that you can knock focus out initially to bring them in range but what does this achieve?
Any help much appreciated.
 
cotty332000 said:
Well I tried this today and wasn't able to get any of my lenses in the +20/-20 range!
Maybe I'm doing something wrong!
They all lost focus in the +range but still had a constant focus confirmation dot at -20.
I am aware that you can knock focus out initially to bring them in range but what does this achieve?
Any help much appreciated.
Click to expand...

Possibly answered in the post above yours?
 
Ive been pondering this - couldnt 'Longitudinal Chromatic Aberration' / 'bokeh fringing' be used to calibrate autofocus. It is a sure way of showing whether an object is in front of or behind the focus plane. All you need is a focus model that produces specular highlights across the focal plane.
 
I 'DotTuned' three of my lenses a couple of months ago and have just bought a LensAlign Pro (2nd hand from the classifieds). Gratified and reassured to find that my results from the LensAlign are almost identical to my DotTune results... I guess it works then!
 
I check my lens at 25 times focal length. I don't use anything fancy. Tried dot tune but just found it a pain. I use 5 high contrast targets with the middle one the zero and the rest at 2 cm intervals in front and behind the zero one. That gives me a dof of 8 cms front to back. Then I take 5 images defocusing each time. I can tell on the screen of the camera how close it is. Only put it on the computer for a final check.

Once I'm happy then I shoot several real world subjects at different angles as a final check.
 
Are you guys using the chart they supply to download or just any plain flat surface for this procedure?
 
OK cheers Dave.

I'll have to look into downloading the one they have on there then as I don't have any other MA chart to use.
 
James Thomas 75 said:
OK cheers Dave.

I'll have to look into downloading the one they have on there then as I don't have any other MA chart to use.
Click to expand...

Did you miss post 23 on this thread? All you need is a box and a ruler ;)
 
HoppyUK said:
Did you miss post 23 on this thread? All you need is a box and a ruler ;)
Click to expand...
I missed all of it so far :D
I've only just found the thread on my mobile so I haven't read through it properly as yet.
I'll be sure to read start to finish when I get home though thanks.
 
hollis_f said:
This post contains a link to the chart I used to use before FoCal.
Click to expand...

Why use an ISO chart? Page from a magazine would be just as good for that method.

And shooting a frame-filler of an A4 page is asking for trouble - it's too small, too close. The only reason you got away with it is because you're using a £5k lens. With a consumer zoom, correcting any errors at that distance will almost certainly put the lens out at more normal range.
 
Tried this with a piece of paper taped to a wall for critical focus, ended up with +10 adjustment on a D600 with sigma 35 1.4

I must say It has improved focus when im after shallow depth of field , worth doing as it only takes a few minutes
 
You must log in or register to reply here.
Back
Top