300mm f/4 lens Image Stabilization oddities.

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,190
Was testing the AF on the 300mm f/4 on the E-M1X. Found a couple of things odd about it.1. The lens has a slight AF Tuning focus shift with the Image Stabilization ON and OFF. It moves about 1 to 1.5 points between the two according to the Focus Align software and using 10 shots at each AF tuning spot. Probably within Olympus specs so I'm not going to worry about it knowing it is there. Charts below where lens IS switch ON the best AF tuning is -1. When switched OFF around 0 to a +1.5 (If a +1.5 were available.). The shot-to-shot AF deviation appears greater with it switched ON too.

2. With the lens IS turned OFF, the thing fires like a machine gun verses when it switched ON and using Low-Speed mechanical Silent Shutter and taking 10 shots at 10 FPS. Sort of startling really!

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

300mm-IS-turned-OFF-10-shots.jpg
Subscribe to see EXIF info for this image (if available)
 
Last edited:

Reflector

Mu-43 Hall of Famer
Joined
Aug 31, 2013
Messages
2,173
This is from the E-M1II menu structure but change this setting:
Gear menu, C2, Image Stabilization
There should be an option for FPS priority or IS priority. You're likely on IS priority which allows the stabilization system to recenter for each shot so it doesn't run out of travel as easily.
 

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,190
  • Thread Starter Thread Starter
  • #3
This is from the E-M1II menu structure but change this setting:
Gear menu, C2, Image Stabilization
There should be an option for FPS priority or IS priority. You're likely on IS priority which allows the stabilization system to recenter for each shot so it doesn't run out of travel as easily.
Thanks!

You are right. It was on IS Priority. Will recheck on FPS Priority and see what happens.
 

Giz9

Mu-43 Rookie
Joined
Mar 9, 2018
Messages
23
Was testing the AF on the 300mm f/4 on the E-M1X. Found a couple of things odd about it.1. The lens has a slight AF Tuning focus shift with the Image Stabilization ON and OFF. It moves about 1 to 1.5 points between the two according to the Focus Align software and using 10 shots at each AF tuning spot. Probably within Olympus specs so I'm not going to worry about it knowing it is there. Charts below where lens IS switch ON the best AF tuning is -1. When switched OFF around 0 to a +1.5 (If a +1.5 were available.). The shot-to-shot AF deviation appears greater with it switched ON too.

2. With the lens IS turned OFF, the thing fires like a machine gun verses when it switched ON and using Low-Speed mechanical Silent Shutter and taking 10 shots at 10 FPS. Sort of startling really!

View attachment 812195

View attachment 812196
That's a very interesting result. I hadn't seen this system and software before. I have struggled to set meaningful values for lens adjustment on the E-M1 Mkii & iii.

One thought occurs to me. Are you using continuous AF or single AF? I see FocusTune says to use S-AF but, on Olympus cameras, this is a hybrid combination of pdaf and cdaf. C-AF only uses pdaf so would seem to be more approriate in this case.

Also a question. What is the units of the left hand axis in the graphs you show? Or are they just some relative measure?

I look forward to hearing if you draw any further conclusions about AF adjustment on Olymous cameras.
 

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,190
  • Thread Starter Thread Starter
  • #5
This is from the E-M1II menu structure but change this setting:
Gear menu, C2, Image Stabilization
There should be an option for FPS priority or IS priority. You're likely on IS priority which allows the stabilization system to recenter for each shot so it doesn't run out of travel as easily.
Better agreement with the IS being ON or OFF today with the switch to set FPS. Both seem to like a -1 AF as the final AF Tuning number. Day was a little overcast as the EV changed a bit. Dunno.

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


I noticed at about 25 feet, a two point change in the AF Tune seems to equate to about a one inch focus shift on the FocusAlign ruler.

As I shoot the hummingbirds at around nine feet, a two point AF focus point change is about 3/8" or so. DOF is already quite shallow at that distance so it's easy to get a missed focus shot. Also, at nine feet mine (annoyingly) seems to front focus a bit verses further out at 25 feet so I often need to apply maybe a +2 to the default. Could be with the small birds the C-AF+Tracking square is maybe four times larger than the non-Tracking C-AF square and might be targeting their closest wing rather than their body, or the tail if the are facing away from me at the feeder and I focus on their head with the smallest AF square yet the tail is in focus and head isn't. I think Sony had that issue with their early Eye Focus where they would lock on the nose tip or brow rather than the deeper eye so they later applied a bit of back focus to the Eye Focus (Helps to have more AF targeting pixels too.).

I know with the Sigma Art lenses in FF size, one can user calibrate with their dock and software maybe 5-6 distance zones for a prime lens and much more for a zoom. Could be the same for the Olympus lenses too where a near will be different from a far distance. Too bad Olympus doesn't offer some AF tuning package as they seem rather mum on the topic.
 

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,190
  • Thread Starter Thread Starter
  • #6
That's a very interesting result. I hadn't seen this system and software before. I have struggled to set meaningful values for lens adjustment on the E-M1 Mkii & iii.

One thought occurs to me. Are you using continuous AF or single AF? I see FocusTune says to use S-AF but, on Olympus cameras, this is a hybrid combination of pdaf and cdaf. C-AF only uses pdaf so would seem to be more approriate in this case.

Also a question. What is the units of the left hand axis in the graphs you show? Or are they just some relative measure?

I look forward to hearing if you draw any further conclusions about AF adjustment on Olymous cameras.
Software only works in C-AF. If you use S-AF, or stab the shutter button too quick and get a S-AF instead of a C-AF (And that sometimes happens randomly between my E-M1X and E-M1 II both for some unknown reason!), the software will warn you of the error. I've learned to fire off a lot more shots at a given tuning point as just one isn't enough and might make it a "Do-it-all-over" event.

The scale on the left is probably some contrast range set by the software as read off the target. It's not user settable.

I wish Olympus would provide some kit to tune these things. Ridiculous to pile in 25 AF points for a Prime and 50 for a Zoom lens. Takes all day to do it. Sigma did it as some cheap FF camera bodies do not allow you to fine tune the AF so you can do it with their hardware dock and software and burn it into the lens firmware. Still a pain though, but better than sending it in to Sigma and praying their tech doesn't get lazy and skip some tuning distance zones (Damhik!).
 

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,190
  • Thread Starter Thread Starter
  • #8
Do you set different values for the AF points? If not, it is quick to just set it for all points.
I've done it for all of them. They are all different to some degree.

My 300mm f/4 AF Tuning results works out to these:

E-M1X AF Tuning Values:
0 1 2 1 1
1 1 -1 1 1
-1 -1 -1 -1 0
1 -1 0 1 1
-3 -3 -4 -2 -2

E-M1 Mark II AF Tuning Values:
2 -1 1 0 0
0 0 0 0 0
-1 0 -2 0 0
-1 0 0 0 0
0 -1 -3 0 -2

The odd -3 and -4 on the bottom edge in the middle seem to agree with the Imaging Resources website's Lab View at f/4 Blur Units too (It cleans up at f/5.6). Don't know what's up with that, but likely normal.

The FocusAlign and it software doesn't work so well with the ruler off to one side and trying to use the AF adjustment zones on the far right in those 25 AF fine-tuning areas. In the middle it is fine and on the left ones, but the ruler positioning doesn't play well with the software as the red indicator dots you need to set are sometimes outside of the image.

I did make up some slant target with the ruler lines on both sides that seemed to work. It gave me the -1 value same as the FocusAlign and its software did. I also used FastRawViewer to confirm the accuracy of the AF Tune for each AF Tuning spot once done. I use that software to cull through all images looking for their sharpness too.

It's in this thread on here: https://www.mu-43.com/threads/em1-micro-focus-adjustment-why-and-how-to-perform.95155/post-1367450
 

Bengeo

Mu-43 Regular
Joined
Feb 13, 2014
Messages
59
Location
Herts, UK
Real Name
Andy
I've done it for all of them. They are all different to some degree.
Interesting, but you have far more patience than me! My lenses have been OK, but I have had to make adjustments for the converters and I've done a single change for all points and that seems to work for my subjects.
 

Giz9

Mu-43 Rookie
Joined
Mar 9, 2018
Messages
23
I've done it for all of them. They are all different to some degree.

My 300mm f/4 AF Tuning results works out to these:

E-M1X AF Tuning Values:
0 1 2 1 1
1 1 -1 1 1
-1 -1 -1 -1 0
1 -1 0 1 1
-3 -3 -4 -2 -2

E-M1 Mark II AF Tuning Values:
2 -1 1 0 0
0 0 0 0 0
-1 0 -2 0 0
-1 0 0 0 0
0 -1 -3 0 -2

The odd -3 and -4 on the bottom edge in the middle seem to agree with the Imaging Resources website's Lab View at f/4 Blur Units too (It cleans up at f/5.6). Don't know what's up with that, but likely normal.

The FocusAlign and it software doesn't work so well with the ruler off to one side and trying to use the AF adjustment zones on the far right in those 25 AF fine-tuning areas. In the middle it is fine and on the left ones, but the ruler positioning doesn't play well with the software as the red indicator dots you need to set are sometimes outside of the image.

I did make up some slant target with the ruler lines on both sides that seemed to work. It gave me the -1 value same as the FocusAlign and its software did. I also used FastRawViewer to confirm the accuracy of the AF Tune for each AF Tuning spot once done. I use that software to cull through all images looking for their sharpness too.

It's in this thread on here: https://www.mu-43.com/threads/em1-micro-focus-adjustment-why-and-how-to-perform.95155/post-1367450
Are these results repeatable? These different measurements could just indicate variations in the C-AF process?

It would be very valuable to know if the results are reapeatable because the process is so time-consuming. I haven't put the time in that you have because I wasn't sure if it was just AF variability.
 

Mack

Mu-43 All-Pro
Joined
Jan 14, 2018
Messages
1,190
  • Thread Starter Thread Starter
  • #11
Are these results repeatable? These different measurements could just indicate variations in the C-AF process?

It would be very valuable to know if the results are reapeatable because the process is so time-consuming. I haven't put the time in that you have because I wasn't sure if it was just AF variability.
They seem so when I later got the idea to double-check the results in FastRawViewer. Note though that the FocusTune software does show some Shot-to-Shot Deviation where you can see the dots on the scatter chart differ although they were each shot ten times at each AF Tuning value. Each shot is not the same. Just demonstrates that although you set in a -1, it doesn't necessarily mean you get the exact same plane of focus each time you trigger a shot with that setting. I've fired off a high-speed burst sequence of a perched bird and could tell the sharpness moved down the branch in FastRawViwer.

On that slant chart I made in the link above, I can see the white lines snap into sharpness near the targets when pressing the shutter. If I move the AF target to another side, same thing occurs and they appear to be centered over the targeting box if the AF Tuning is correct. One can change the tuning number and observe the lines move up with a plus setting, and move down by changing it to a minus setting. Double check afterwards in FastRawViewer if you are uncertain you entered the correct value to center them or not, that gets rid of any personal subjective idea.

The lens does wander a small amount and best we can do is find its best average to set into it. It does seem odd Olympus made so many AF Tuning points where other manufactures, e.g. Nikon and Canon, only allow for one or two AF Tuning spots. However, the 25 Olympus does provide do change the overall sharpness as seen around the sensor surface. My concern is shooting some BIF and if flies into some other AF Tuning area, might that affect the sharpness had it not been the same from the AF Tuning spot it started in.

( I really shouldn't ponder this stuff while waiting for the **** birds to show up! :popcorm2: )
 
Last edited:

Giz9

Mu-43 Rookie
Joined
Mar 9, 2018
Messages
23
They seem so when I later got the idea to double-check the results in FastRawViewer. Note though that the FocusTune software does show some Shot-toShot Deviation where you can see the dots on the scatter chart differ although they were each shot ten times at each AF Tuning value. Each shot is not the same. Just demonstrates that although you set in a -1, it doesn't necessarily mean you get the exact same plane of focus each time you trigger a shot with that setting. I've fired off a high-speed burst sequence of a perched bird and could tell the sharpness moved down the branch in FastRawViwer.

On that slant chart I made in the link above, I can see the white lines snap into sharpness near the targets when pressing the shutter. If I move the AF target to another side, same thing occurs and they appear to be centered over the targeting box if the AF Tuning is correct. One can change the tuning number and observe the lines move up with a plus setting, and move down by changing it to a minus setting. Double check afterwards in FastRawViewer if you are uncertain you entered the correct value to center them or not, that gets rid of any personal subjective idea.

The lens does wander a small amount and best we can do is find its best average to set into it. It does seem odd Olympus made so many AF Tuning points where other manufactures, e.g. Nikon and Canon, only allow for one or two AF Tuning spots. However, the 25 Olympus does provide do change the overall sharpness as seen around the sensor surface. My concern is shooting some BIF and if flies into some other AF Tuning area, might that affect the sharpness had it not been the same from the AF Tuning spot it started in.

( I really shouldn't ponder this stuff while waiting for the **** birds to show up! :popcorm2: )
Thanks for the reply. Maybe I should invest in the target and software. Where did you get your copy of LensAlign/FocusTune?
 
Last edited:

Latest posts

Links on this page may be to our affiliates. Sales through affiliate links may benefit this site.
Mu-43 is a fan site and not associated with Olympus, Panasonic, or other manufacturers mentioned on this site.
Forum post reactions by Twemoji: https://github.com/twitter/twemoji
Copyright © 2009-2019 Amin Forums, LLC
Top Bottom