mplkn
Forum Enthusiast
Ok I'll dig back through the thread: were the two 45s the same?
-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Fuji GFX 100S focus bracketing doesn't return to starting position reliably
- Thread starter JimKasson
- Start date
Last edited:
Hamiltionian
Senior Member
Not bad at all indeed. Significantly better than the 110mm.
Not quite as good as the 63, but not bad at all.
--
Stay Calm and Carry Cameras
Indeed. You may have noticed that now that I'm doing this for every lens, I'm using Matlab for the graphs instead of Excel. Much less labor intensive.Not bad at all indeed. Significantly better than the 110mm.
Not quite as good as the 63, but not bad at all.
--
https://blog.kasson.com
Last edited:
Hamiltionian
Senior Member
Yeah, once you get the scripts set up, way easier. I'll be curious to see how the other lenses turn out. My guess is that most of the others should be pretty good with the possible exceptions of the 120 where I saw some of the same jitter behavior as with the 110.
--
the last word the last word - Photography meets digital computer technology. Photography wins -- most of the time.
Photography meets digital computer technology. Photography wins -- most of the time.
blog.kasson.com
Erik Kaffehr
Veteran Member
Can you see common traits between 'good' and 'less good' lenses with regards to focus shift/focus slip?
Best regards
Erik
--
Erik Kaffehr
Website: http://echophoto.dnsalias.net
Magic uses to disappear in controlled experiments…
Gallery: http://echophoto.smugmug.com
Articles: http://echophoto.dnsalias.net/ekr/index.php/photoarticles
Can you see common traits between 'good' and 'less good' lenses with regards to focus shift/focus slip?
Too much randomness for me to tell.
--
the last word the last word - Photography meets digital computer technology. Photography wins -- most of the time.
Photography meets digital computer technology. Photography wins -- most of the time.
blog.kasson.com
Erik Kaffehr
Veteran Member
Just to say, it's good to see that most (?) lenses may not have the issue.Too much randomness for me to tell.Can you see common traits between 'good' and 'less good' lenses with regards to focus shift/focus slip?
Also, I would think that the issue interferes with your testing, but it would not matter a lot in 'real' photography.
Best regards
Erik
--
Erik Kaffehr
Website: http://echophoto.dnsalias.net
Magic uses to disappear in controlled experiments…
Gallery: http://echophoto.smugmug.com
Articles: http://echophoto.dnsalias.net/ekr/index.php/photoarticles
15 series. Shift after first one, then all the same.
--
blog.kasson.com
--
the last word the last word - Photography meets digital computer technology. Photography wins -- most of the time.
Photography meets digital computer technology. Photography wins -- most of the time.
blog.kasson.com
Hamiltionian
Senior Member
Well that is frustrating. At least with the 110mm you know you can't trust the focus shift data due to systematic error. This is so close.15 series. Shift after first one, then all the same.
--
Stay Calm and Carry Cameras
But weirdly consistent.
--
the last word the last word - Photography meets digital computer technology. Photography wins -- most of the time.
Photography meets digital computer technology. Photography wins -- most of the time.
blog.kasson.com
Wow, I wasn't expecting that given that the other lenses which tested poorly were larger aperture lenses. Bummer.
But weirdly consistent.
The 120 starts out drifting:
But then settles down, albeit with a glitch at the beginning:
--
blog.kasson.com
But then settles down, albeit with a glitch at the beginning:
--
the last word the last word - Photography meets digital computer technology. Photography wins -- most of the time.
Photography meets digital computer technology. Photography wins -- most of the time.
blog.kasson.com
Jack Hogan
Veteran Member
Unusual, the first three sets it never hit peak 'focus', is there something else going on there?The 120 starts out drifting:
But then settles down, albeit with a glitch at the beginning:
I dunno. Maybe vibration for the first series. I use a 10-second self timer when I'm doing the f-stop series, but not for these focus bracketing series. As to the other ones, maybe there's some residual vibration from the camera's racking focus back to what it thinks is the original position.Unusual, the first three sets it never hit peak 'focus', is there something else going on there?The 120 starts out drifting:
But then settles down, albeit with a glitch at the beginning:
--
the last word the last word - Photography meets digital computer technology. Photography wins -- most of the time.
Photography meets digital computer technology. Photography wins -- most of the time.
blog.kasson.com
mplkn
Forum Enthusiast
I've been sitting here watching this thread thinking about how the system works and realizing I've never read a thorough description.
There's firmware in the lens, living between the contacts and the focusing motor, taking instructions from the brain and relaying positional information back to the brain (in the camera body), does that sound right?
So, a little mechanical slop in the physical focus mechanism of the individual lens example could result in a difference between the lens's reported position and it's actual position, right?
So the brain might be happy, because the lens is reporting all's well, while in actuality its actual physical focus position would be something else, varying, falling somewhere inside the range of error allowed by the physical slop.
I'm totally making this up. I don't know in detail how the tasks are divided between lens and body. I'm going with assumptions based on some guesses as to how the control system works.
If there's any truth to these guesses, however, it might mean that some examples of a given lens might be better than others. I guess I don't think this is a camera body issue; my guess is the body and brain are doing the best they can based on the information they're getting from the lens.
Thoughts?
There's firmware in the lens, living between the contacts and the focusing motor, taking instructions from the brain and relaying positional information back to the brain (in the camera body), does that sound right?
So, a little mechanical slop in the physical focus mechanism of the individual lens example could result in a difference between the lens's reported position and it's actual position, right?
So the brain might be happy, because the lens is reporting all's well, while in actuality its actual physical focus position would be something else, varying, falling somewhere inside the range of error allowed by the physical slop.
I'm totally making this up. I don't know in detail how the tasks are divided between lens and body. I'm going with assumptions based on some guesses as to how the control system works.
If there's any truth to these guesses, however, it might mean that some examples of a given lens might be better than others. I guess I don't think this is a camera body issue; my guess is the body and brain are doing the best they can based on the information they're getting from the lens.
Thoughts?
Erik Kaffehr
Veteran Member
I would think that there is a focusing motor and a position sensor. There may also be some mechanical play between focusing motor and focusing group involved.
How these things work together, who knows?
Best regards
Erik
It's a control system. Part of the system is in the camera, and part is in the lens. They both are necessary to make focus bracketing work. Usually, in voice coil actuator systems, if absolute positioning is necessary, there is an absolute or relative position sensor. If relative, the control system needs to return the mechanism to a known position (such as against one of the travel stops), then compute absolute position from there. Position sensors are subject to lack of precision, absolute accuracy errors, relative accuracy errors, hysteresis, backlash, etc. So are VCA control systems. If the lens is focused by as stepper motor, that has its own set of precision and accuracy issues. The system computing the stimulus to the motor is also subject to precision constraints, and in addition the time to complete the computations may depend on what else is going on in the camera.
I don't think that, in the absence of a teardown, it's particularly fruitful to try to find out where the focus bracketing ball is being dropped. It's enough to know that it sometimes is, and that it varies with lens model. It will be useful to see if it varies by lens serial number, if the model is the same. Once we understand the limitations of the system, taken as a block box, we can devise workarounds.
mplkn
Forum Enthusiast
Yes, thanks, Jim. I was thinking steppers could offer positional accuracy, as each position, in effect, has an address. But that still leaves wide open all the other potential sources of untoward behavior, such as you've listed. That was a great list, by the way, all of which are elements of the eternal, verifiable, general cussedness of contraptions.
I agree it'll be useful to know if the behavior varies by serial number, that is, by individual example: or at least, interesting.
Does anyone have a sense of whether Fujifilm's stuff is particularly egregious compared to other manufacturers? I'd hazard a guess this trait might not be unique to Fujifilm...
I agree it'll be useful to know if the behavior varies by serial number, that is, by individual example: or at least, interesting.
Does anyone have a sense of whether Fujifilm's stuff is particularly egregious compared to other manufacturers? I'd hazard a guess this trait might not be unique to Fujifilm...
Last edited:
Keyboard shortcuts
- f
- Forum