Ugly Hedgehog - Photography Forum
Home Active Topics Newest Pictures Search Login Register
Main Photography Discussion
Reikan FoCal Pro Lens Calibration
Page <prev 2 of 2
Nov 23, 2022 10:29:02   #
abc1234 Loc: Elk Grove Village, Illinois
 
User ID wrote:
That challenges common sense.

Heres how it works in various cameras that I use:

The lens manufacturer calibrates nothing, but does provide that lens element can move when commanded to move by the body AF drive.

The body AF drive is informed by the sensor imagedata as to which way the elementsshould move, and as to when the morion must cease.

Nothing involved there needs "calibration".


As you said, that challenges common sense. If my explanation is wrong, then I will admit to that. Nonetheless, I have found that my lenses improved by recalibration, also known as microadjustment. One can certainly question whether the improvement is noticeable or have lenses, new or used, that are optimal. I do accept both of these assertions. For me, I prefer to assume nothing and to conduct carefully controlled experiments.

Reply
Nov 23, 2022 11:43:08   #
Architect1776 Loc: In my mind
 
abc1234 wrote:
Without knowing the experimental details and seeing the results, I would hesitate on generalizing from your experience.


I could care less.
That was with my cameras and lenses.
You for all I know you have defective equipment and need to try to fix it. Good for you.
But that is your world. I gave a real world experience and for me that is sufficient.

Reply
Nov 23, 2022 12:03:58   #
User ID
 
abc1234 wrote:
As you said, that challenges common sense. If my explanation is wrong, then I will admit to that. Nonetheless, I have found that my lenses improved by recalibration, also known as microadjustment. One can certainly question whether the improvement is noticeable or have lenses, new or used, that are optimal. I do accept both of these assertions. For me, I prefer to assume nothing and to conduct carefully controlled experiments.

If your in-out focus data is gathered from the image sensor then theres no reason for focus errors. OTOH, if its gathered from some other source such as a Leica rangefinder or an AF SLRs AF module, those other sources are only analogous to the real image focused on the sensor. There is artifice involved and thus a possible difference between the analog and the reality.

Any AF calibration performed will be to deal with the difference between the non-sensor focusing device (Leica RF or SLR AF module) and the actual sensor image. IOW, you are calibrating the bodys own internal focusing device, but nothing in the lens.

The only focusing info that comes from an SLR lens is the light beams coming thru the glass. Theres nothing within the lens to calibrate. Its the bodys assigned job to determine when those light beams are in focus. The body will ask the lens to move, and when the body sees in-focus it tells the lens to stop moving. Any inaccuracy there is the bodys error.

(With Leicas the lenses are also calibrated cuz the RF is just a mechanical position measuring device, it doest see any light from the lens).

Reply
 
 
Nov 23, 2022 12:36:24   #
abc1234 Loc: Elk Grove Village, Illinois
 
User ID wrote:
If your in-out focus data is gathered from the image sensor then theres no reason for focus errors. OTOH, if its gathered from some other source such as a Leica rangefinder or an AF SLRs AF module, those other sources are only analogous to the real image focused on the sensor. There is artifice involved and thus a possible difference between the analog and the reality.

Any AF calibration performed will be to deal with the difference between the non-sensor focusing device (Leica RF or SLR AF module) and the actual sensor image. IOW, you are calibrating the bodys own internal focusing device, but nothing in the lens.

The only focusing info that comes from an SLR lens is the light beams coming thru the glass. Theres nothing within the lens to calibrate. Its the bodys assigned job to determine when those light beams are in focus. The body will ask the lens to move, and when the body sees in-focus it tells the lens to stop moving. Any inaccuracy there is the bodys error.

(With Leicas the lenses are also calibrated cuz the RF is just a mechanical position measuring device, it doest see any light from the lens).
If your in-out focus data is gathered from the ima... (show quote)


Thank you for the clarification.

Reply
Nov 23, 2022 12:54:17   #
TriX Loc: Raleigh, NC
 
This is like the Mac vs PC argument – some people (myself included) cal all their DSLR lenses and a substantial proportion of professionals do it as a matter of course, while others, like Bill, don’t do it (and have likely never tried it) and produce professional results.

Here’s the simple explanation: with both the lens and the camera, every dimension in the assembly has a tolerance when machined and assembled. And to pass QA, the tolerance stack up of a lens or camera must be within the assembled tolerance. BUT, what happens when the camera assembly is at the far negative end of the allowable tolerance and the lens it’s at the far positive end of the acceptable tolerance? Errors in autofocus. To address this, higher end and pro cameras have a microfocus adjustment capability - typically one for a prime and two for a zoom. The camera stores the calibration for each lens (by type and S/N) and applies it when that lens is mounted, making the AF close to exact.

The question (if you choose to calibrate and get 100% of the sharpness you paid big $ for) is how to do it accurately. There are a number of methods, but most except FoCal require subjective judgement. Focal uses a target and defocuses then refocuses at a number of points using different calibrations to arrive at an accurate graph of acuity vs microfocus adjustment and inputs the best one for each lens (or two for a zoom) into the camera’s memory. Focal also tests your lens for acuity at each aperture, so you KNOW the sharpest aperture, when diffraction becomes noticeable, and also identifies sub par lenses. I find it a great tool and worth every penny.

Now as to the argument(s) against it (usually by those that have never tried it), they are typically twofold. First, they’ve been shooting good sharp shots for years without calibration. To which I respond: (a) if you haven’t tried it and compared, how do you know if your images couldn’t be sharper if you used it? (b) if you typically shoot a few stops down, you may never notice it because the DOF “hides it”, but if you shoot wide open where the DOF may be 6”, having the AF off by a few inches can make a BIG difference. The second argument is that it’s only correct a one distance or at most, 2 FLs for a zoom. I’ve tested this and posted the results a few years ago on UHH, and the conclusion was that while the MFA did vary by both distance and FL for zooms, it was ALWAYS better than no adjustment. Look up the post if you’re interested.

To sum up, I’ve proven to myself with extensive testing, that most lenses need some adjustment and one went from being unacceptable to very useful. You can absolutely see the difference, but you’ll never get it right with angled rulers or just randomly trying some settings. But if you use Focal, you’ll KNOW that $1500 lens is giving you everything you paid for instead of performing like a $300 lens when you shoot wide open. And BTW, mirrorless cameras don’t typically need calibration.

Reply
Nov 23, 2022 13:40:06   #
User ID
 
TriX wrote:
This is like the Mac vs PC argument – some people (myself included) cal all their DSLR lenses and a substantial proportion of professionals do it as a matter of course, while others, like Bill, don’t do it (and have likely never tried it) and produce professional results.

Here’s the simple explanation: with both the lens and the camera, every dimension in the assembly has a tolerance when machined and assembled. And to pass QA, the tolerance stack up of a lens or camera must be within the assembled tolerance. BUT, what happens when the camera assembly is at the far negative end of the allowable tolerance and the lens it’s at the far positive end of the acceptable tolerance? Errors in autofocus. To address this, higher end and pro cameras have a microfocus adjustment capability - typically one for a prime and two for a zoom. The camera stores the calibration for each lens (by type and S/N) and applies it when that lens is mounted, making the AF close to exact.

The question (if you choose to calibrate and get 100% of the sharpness you paid big $ for) is how to do it accurately. There are a number of methods, but most except FoCal require subjective judgement. Focal uses a target and defocuses then refocuses at a number of points using different calibrations to arrive at an accurate graph of acuity vs microfocus adjustment and inputs the best one for each lens (or two for a zoom) into the camera’s memory. Focal also tests your lens for acuity at each aperture, so you KNOW the sharpest aperture, when diffraction becomes noticeable, and also identifies sub par lenses. I find it a great tool and worth every penny.

Now as to the argument(s) against it (usually by those that have never tried it), they are typically twofold. First, they’ve been shooting good sharp shots for years without calibration. To which I respond: (a) if you haven’t tried it and compared, how do you know if your images couldn’t be sharper if you used it? (b) if you typically shoot a few stops down, you may never notice it because the DOF “hides it”, but if you shoot wide open where the DOF may be 6”, having the AF off by a few inches can make a BIG difference. The second argument is that it’s only correct a one distance or at most, 2 FLs for a zoom. I’ve tested this and posted the results a few years ago on UHH, and the conclusion was that while the MFA did vary by both distance and FL for zooms, it was ALWAYS better than no adjustment. Look up the post if you’re interested.

To sum up, I’ve proven to myself with extensive testing, that most lenses need some adjustment and one went from being unacceptable to very useful. You can absolutely see the difference, but you’ll never get it right with angled rulers or just randomly trying some settings. But if you use Focal, you’ll KNOW that $1500 lens is giving you everything you paid for instead of performing like a $300 lens when you shoot wide open. And BTW, mirrorless cameras don’t typically need calibration.
This is like the Mac vs PC argument – some people ... (show quote)

Its very much like how lens makers dont hafta tediously accurately position the infinity focus hard stop if we can all just accept that its perfectly normal for all lenses to focus beyond infinity.

Reply
Nov 23, 2022 22:12:22   #
abc1234 Loc: Elk Grove Village, Illinois
 
I recommend that everyone read TriX's post very carefully. The only side he takes is that of objectivity. Thanks for taking the time to formulate this informative post.

Reply
 
 
Nov 23, 2022 22:44:28   #
TriX Loc: Raleigh, NC
 
abc1234 wrote:
I recommend that everyone read TriX's post very carefully. The only side he takes is that of objectivity. Thanks for taking the time to formulate this informative post.


Thank you for the kind remarks. Here are two threads, one testing the contention that the microfocus adjustment (MFA) isn’t useful because it’s only correct at one FL: https://www.uglyhedgehog.com/t-508300-1.html.

And another showing Focal’s facility to plot aperture vs acuity (and some results that are contrary to the idea that the sharpest aperture is always 1-2 stops down from wide open). Even though MFA and Focal isn’t applicable to MILC cameras, I wish they’d make this utility available for them - it tells you in a minute if that lens you just bought is up to snuff, where it’s really the sharpest, and when diffraction begins to have a noticeable effect: https://www.uglyhedgehog.com/t-573232-1.html

Reply
Nov 23, 2022 22:56:00   #
abc1234 Loc: Elk Grove Village, Illinois
 
TriX wrote:
Thank you for the kind remarks. Here are two threads, one testing the contention that the microfocus adjustment (MFA) isn’t useful because it’s only correct at one FL: https://www.uglyhedgehog.com/t-508300-1.html.

And another showing Focal’s facility to plot aperture vs acuity (and some results that are contrary to the idea that the sharpest aperture is always 1-2 stops down from wide open). Even though MFA and Focal isn’t applicable to MILC cameras, I wish they’d make this utility available for them - it tells you in a minute if that lens you just bought is up to snuff, where it’s really the sharpest, and when diffraction begins to have a noticeable effect: https://www.uglyhedgehog.com/t-573232-1.html
Thank you for the kind remarks. Here are two threa... (show quote)


You are welcome. Of the four lenses I tested, all were sharpest at about two stops down. The falloff was not steep so you not notice much of difference. However, the big take away is you really do not know until you test them properly. FoCal sure beats those cheap, static, and subjective tests.

Reply
Nov 23, 2022 23:26:11   #
jcboy3
 
billnikon wrote:
Many experts caution against fine tuning a lens, or, if you will, lens calibration. Most calibrations can only be done for one specific distance, many, many folks who try to fine turn end worse off than before calibration.
Over my 50 years of professional photography I have never had to fine tune a lens, be it Canon, Nikon, Sony, or Hasselblad.
I now only shoot mirrorless, which focuses the image right on the sensor, no need for any calibration for any lens.
I caution anyone trying to calibrate, DO NOT DO IT.
Lots of folks here will disagree with my statement, but after 50 years of hanging photography, and after more than 500 weddings, I have never, NEVER, had any issues with not getting a tack sharp image.
If you know what your doing, you will NEVER have to fine turn a lens.
Goodbye.
Many experts caution against fine tuning a lens, o... (show quote)


Some mirrorless cameras allow for calibration of the focus points, which can vary in accuracy when the camera is using PDAF focusing. Testing this can be tedious with higher end cameras, which can have a lot of focus points. The need is due to the design of the PDAF sites on the image sensor. But if you are getting variable focus with continuous AF, testing might reveal PDAF site variance to be the reason.

As for DSLRs, I have calibrated my telephoto lenses and some of the lenses were sufficiently off to warrant calibration. Calibration should be done at distances that the lens is used. I calibrated at multiple distances and the adjusted to average the overall error.

I initially used Reikan FoCal for my Nikon DSLRs, but got similar results more quickly using the green dot method. So I stopped updating the software. I have not looked into testing my mirrorless cameras yet.

Reply
Nov 24, 2022 00:49:13   #
User ID
 
abc1234 wrote:
I recommend that everyone read TriX's post very carefully. The only side he takes is that of objectivity. Thanks for taking the time to formulate this informative post.

I saw it, but it looked verrrry long ... but at a quick glance it does look about right.
;-)

Reply
Page <prev 2 of 2
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.
Main Photography Discussion
UglyHedgehog.com - Forum
Copyright 2011-2024 Ugly Hedgehog, Inc.